NiNth EditioNGlobal EditioN Qualitative Research Methods for the Social Sciences Howard Lune Hunter College, CUNY Bruce L... Preface 9 1.1: Qualitative Methods, Qualitative Data 12 1.2:
Trang 2NiNth EditioN
Global EditioN
Qualitative Research Methods for the
Social Sciences
Howard Lune
Hunter College, CUNY
Bruce L Berg
California State University, Long Beach
Harlow, England • London • New York • Boston • San Francisco • Toronto • Sydney • Dubai • Singapore • Hong Kong
Tokyo • Seoul • Taipei • New Delhi • Cape Town • Sao Paulo • Mexico City • Madrid • Amsterdam • Munich • Paris • Milan
www.downloadslide.com
Trang 3VP, Product Development: Dickson Musslewhite
Director, Content Strategy and Development:
Sharon Geary
Editor in Chief: Ashley Dodge
Managing Editor: Sutapa Mukherjee
Sponsoring Editor: Bimbabati Sen
Content Manager: Carly Czech
Project Manager, Global Edition: Sudipto Roy
Senior Acquisitions Editor, Global Edition:
Sandhya Ghoshal
Senior Project Editor, Global Edition: Daniel Luiz
Project Editor, Global Edition: Rahul Arora
Manager, Media Production, Global Edition:
M Vikram Kumar
Manufacturing Controller, Production, Global
Edition: Jerry Kataria
Editorial Project Manager: Lindsay Bethoney,
Lumina Datamatics, Inc
Asset Development Team: LearningMate
Operations Specialist: Mary Ann Gloriande Associate Director of Design: Blair Brown Interior Design: Kathryn Foot
Cover Design: Lumina Datamatics, Inc.
Cover Art: Westend61 Premium/Shutterstock Full-Service Project Management and Composition:
Ashwina Ragounath/Integra Software Services
Acknowledgements of third party content appear on pages 235–237, which constitutes an extension of this copyright page
Pearson Education Limited
Edinburgh Gate
Harlow
Essex CM20 2JE
England
and Associated Companies throughout the world
Visit us on the World Wide Web at:
www.pearsonglobaleditions.com
© Pearson Education Limited 2017
The rights of Howard Lune and Bruce L Berg to be identified as the authors of this work have been asserted
by them in accordance with the Copyright, Designs and Patents Act 1988
Authorized adaptation from the United States edition, entitled Qualitative Research Methods for the Social Sciences, 9th edition, ISBN 978-0-134-20213-6, by Howard Lune and Bruce L Berg, published by Pearson Education © 2017.
All rights reserved No part of this publication may be reproduced, stored in a retrieval system, or transmitted
in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without either the prior written permission of the publisher or a license permitting restricted copying in the United Kingdom issued by the Copyright Licensing Agency Ltd, Saffron House, 6–10 Kirby Street, London EC1N 8TS
All trademarks used herein are the property of their respective owners The use of any trademark in this text does not vest in the author or publisher any trademark ownership rights in such trademarks, nor does the use
of such trademarks imply any affiliation with or endorsement of this book by such owners
ISBN 10: 129-2-16439-5
ISBN 13: 978-1-292-16439-7
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
10 9 8 7 6 5 4 3 2 1
14 13 12 11 10
Printed and bound in Vivar, Malaysia
Trang 41 Introduction 11
2 Designing Qualitative Research 22
4 A Dramaturgical Look
6 Ethnographic Field Strategies 107
7 Participatory Action Research 136
8 Unobtrusive Measures in Research 146
9 Social Historical Research
Trang 5This page intentionally left blank
Trang 6Preface 9
1.1: Qualitative Methods, Qualitative Data 12
1.2: Use of Triangulation in Research Methodology 14
1.3: Qualitative Strategies: Defining an Orientation 15
1.4: From a Symbolic Interactionist Perspective 17
2.5: Operationalization and Conceptualization 31
2.6.4: Setting and Population Appropriateness 36
2.6.6: Representativeness 39
2.8: Data Storage, Retrieval, and Analysis 40
2.9: Dissemination 41
3.1: Research Ethics in Historical Perspective 44
3.1.1: Regulations in the Research Process 45
3.2: Informed Consent and Implied Consent 46
3.3.2: Strategies for Safeguarding Confidentiality 48
3.6.3: Active versus Passive Consent 553.6.4: Active versus Passive Consent in
3.10: Objectivity and Careful Research Design 62
4 A Dramaturgical Look at Interviewing 65
4.3.2: The Unstandardized Interview 684.3.3: The Semistandardized Interview 69
Contents
www.downloadslide.com
Trang 74.12: Conducting an Interview: A Natural or
4.13.1: Interviewer Roles and Rapport 82
4.13.2: The Role of the Interviewee 83
4.13.3: The Interviewer as a Self-Conscious
Performer 83
4.13.4: Social Interpretations and the Interviewer 84
4.14.1: Interviewers’ Attitudes and Persuading
4.14.2: Developing an Interviewer Repertoire 87
4.14.4: Taking the Show on the Road 88
4.14.5: The Ten Commandments of Interviewing 88
4.16.3: Analysis Procedures: A Concluding
Remark 92
5.2.2: Introduction and Introductory Activities 96
5.2.3: Statement of the Basic Rules or Guidelines
5.2.4: Short Question-and-Answer Discussions 97
5.2.5: Special Activities or Exercises 97
5.2.6: Guidance for Dealing with Sensitive Issues 97
5.4: Selecting Focus Groups as a Method 100
5.7: Common Missteps When Using Focus Groups 103
5.8: Confidentiality and Focus Group Interviews 104
6.1: Accessing a Field Setting: Getting In 109
6.1.1: Negotiating the Researcher’s Role 112
6.2.2: Other Dangers During Ethnographic Research 115
6.3.1: How to Learn: What to Watch and
6.5: Other Analysis Strategies: Typologies,
6.5.1: Typologies 1266.5.2: Sociograms 1276.5.3: Metaphors 129
6.8.1: The Attitude of the Ethnographer 132
7.2: Identifying the Research Question(s) 139
7.4: Analyzing and Interpreting the Information 1397.4.1: Descriptive Accounts and Reports 1407.5: Sharing the Results with the Participants 1407.6: When to Use and When Not to Use Action
Research 141
7.8.1: Technical/Scientific/Collaborative Mode 1427.8.2: A Practical/ Mutual Collaborative/
7.8.3: Emancipating or Empowering/
Enhancing/Critical Science Mode 142
Trang 88.1.3: A Last Remark About Archival Records 154
8.2: Physical Erosion and Accretion: Human Traces
9.2: Life Histories and Social History 160
9.3: What Are the Sources of Data for Historical
9.5.1: Oral History as Reality Check 166
10.3.1: The Use of Personal Documents 174
10.4: Intrinsic, Instrumental, and Collective Case Studies 175
10.6: The Scientific Benefit of Case Studies 177
10.6.1: Objectivity and the Case Method 177
10.6.2: Generalizability 177
10.8.1: Data Collection for Community Case Studies 179
11 An Introduction to Content Analysis 181
11.2.2: Social Anthropological Approaches 18311.2.3: Collaborative Social Research Approaches 18311.2.4: Content Analysis and Theory 18311.3: Content Analysis as a Research Technique 18411.3.1: Quantitative or Qualitative? 18611.3.2: Manifest versus Latent Content Analysis 186
11.4.1: Levels and Units of Analysis 188
12.2: Identifying the Purpose of the Writing 20412.3: Delineating a Supportive Structure: Visual
12.3.2: Original Contribution Sections 208
12.3.4: Discussion/Conclusion 20912.3.5: References, Notes, and Appendices 210
www.downloadslide.com
Trang 912.5: Presenting Research Material 212
12.5.1: Disseminating the Research: Professional
12.6: A Word About the Content of Papers and Articles 215
12.7: Write It, Rewrite It, Then Write It Again! 215
References 219Credits 235
Trang 10Social research provides necessary support for
innumer-able professions, bolsters and directs policy decisions,
fact-checks both wild and mundane claims about the
world, and helps us understand ourselves and others But
even beyond these valuable endeavors, social research has a
simple mission “to help us know what’s going on.” In this
era of what is sometimes called globalization, everyone’s
lives are impacted by vast numbers of things happening all
over the planet, in all segments of industry, society, politics,
economics, culture, and religion Even the well-informed
have little idea about most of it We cannot observe and
un-derstand everything we need on our own Research
com-presses the vast variability of life into more or less consistent
and predictable bits of reality It gives us a leg to stand on
New to the Edition
The new edition of Qualitative Research Methods for the Social
Sciences continues the mission of the original—to teach
stu-dents where our data comes from, how to manage it, how to
make sense of it, what it can mean, and what it can do In this
edition, I have also added an emphasis on the other side of
that coin Each chapter briefly highlights the limitations on
the various methods of data collection and analysis There
are things that research cannot do Well-planned studies with
reliable data and valid analyses can teach us a great deal, but
they are not magic As students of research, we must be
criti-cal consumers as well as producers We have to know where
to set the limits on our own ambitions and how to critically
evaluate the claims that others make based on their
under-standings of the measurable world
Research methods continue to grow and develop in
exciting new ways, through experience, interdisciplinary
conversation, new technologies, and in response to new
needs It has been centuries since maps were routinely
produced with large areas of unknown topology The
world is no longer a mystery of undiscovered places and
people Now we are living with the opposite challenge:
There is too much data Everything we do seems to occur
in public, in measurable ways We are data With
increas-ing use of surveillance technologies, the very concept of
anonymity is losing meaning And, of course, with our
mini-oracles in our pockets ready to search the world’s
databases in less than one second to immediately retrieve
even the most obscure bits of cultural trivia, it seems as
though everything is knowable It isn’t Factoids of
infor-mation, traces of personal histories, photographs, song
lyrics, and train schedules, as well as body counts and
temperature readings are merely data points None of this
is useful information until it is organized, explored, and interpreted Research methods grow to manage larger pools of more diverse data Yet the basic principles and underlying practices remain the same While this text cov-ers both new and old tricks and techniques, my primary purpose is to emphasize the logic of research planning and the elusive task of finding meaning The organization
of chapters and topics remains unchanged since the last edition Our job remains the same
This edition of the book builds on the foundation of the previous editions while offering a number of improve-ments I have corrected errors wherever I could find them and sought to clarify the most confusing discussions I have added new and more challenging exercises and questions for discussion The present edition gives more attention
to visual and spatial analysis and to qualitative analysis software, but only in relation to the familiar methodolo-gies where those tools apply In addition to the challenge of presenting contemporary technologies before they change again, I have updated many of the examples used through-out the book to provide more contemporary data, except in the cases of certain classic studies or exemplary discussions that, to me, are irreplaceable I have also reorganized sec-tions for students in order to provide more clarity and to improve readability
This ninth edition contains expanded discussions in key areas, such as research design, research ethics, and writing
I have given more attention to the context for the different techniques, with explicit attention to when they work best
or least And, to accommodate this new material, I have ciously removed portions of the text throughout Overall, I have tried to serve the two goals that have always driven this text from its first edition: to be as useful and challenging as possible without being dull
judi-This edition of Qualitative Research Methods for the Social Sciences may be read straight through, at approximately one
chapter per week, for 12–15 weeks Or, one can read tively and in any order Each chapter is intended to be suf-ficiently self-contained to allow students to start anywhere and to proceed at your own pace The coverage of materials
selec-is intended to be thorough enough to use as a stand-alone text, while sections are divided in a manner to allow instruc-tors to isolate specific units in conjunction with other texts or readers Most importantly, the advice and exercises offered here are intended to support students’ efforts to actually get out of the classroom and try some of this out There is no better learning method than to throw yourself into it, make mistakes, and figure out what went wrong Success is useful too, but failure can be the best teacher
Preface
www.downloadslide.com
Trang 11Available Instructor Resources
The following instructor resources can be accessed by
visiting http://www.pearsonglobaleditions.com/lune
• Instructor Manual
Detailed instructor’s manual with learning objectives,
chapter outlines, discussion questions, activities, and
assignments
• PowerPoint Presentation
Provides a core template of the content covered
through-out the text; can easily be added to customize for your
of the content, Pearson would like to thank Dave Centeno, University of the Philippines; Timothy Lynch, Plymouth University; and Sanjukta Bhattacharya
Trang 12How do we know things?
Let’s consider a few propositions First, whatever you
think you know about the world is incomplete and likely
to be at least partially wrong Second, experience is a great
teacher, but your experiences probably don’t reflect other
people’s experience of the world all that well And besides,
we are all rather selective about what things we remember
and what lessons we learn from them So even the things
we know from our own lives are somewhat suspect, let
alone things we’ve learned from others It turns out that
this is not a bad thing, as long as we deal with it
realisti-cally But it does not give us a reliable or detailed
under-standing of our society or much beyond it
Cynics can deny the things they don’t like to believe
by asking, “How do you know? Were you there?” This
approach gives the false impression that you can only
know something by direct experience How do people
born after 1969 know that the moon landing wasn’t just
a TV show? How do people who watched it on TV at the
time know that it wasn’t a giant fake produced in Area 51?
Why should I believe in Denmark? I’ve never been there
And if you want to be really difficult with people, you
can always remind them that Plato said that we could
be lying in a cave somewhere cut off from real sensory
input, attached to some kind of matrix-like virtual reality
generator Nonetheless, barring the possibility that the
whole apparent world only exists within a conspiracy
designed to mess with your head, we can proceed with the
assumption that the world is real, observable, and able The “how do you know” question comes down to three parts: What do we observe? How do we measure things? And how is reliable knowledge distinguished from things we are less sure of?
measur-In this book, we’re only going to address these tions for matters of social scientific research I will leave Denmark to some other writer
ques-We’ll start by distinguishing between the social world and the rest From where I sit when I’m writing, I can see mountains in the distance, or I would if I went outside These are observable and real artifacts of the physical world, and therefore not particularly sociological But all the things around them—from the roads that I drive on over the mountain to get to town, to the radio stations that fade in and out depending on which side I’m on or how high up, or the differences between where there are street lights and where darkness, or the politics and economics
of maintaining the reservoir here that provides drinking water elsewhere, or the availability of WiFi in some coffee shops where I write but not in others where I don’t—are all artifacts of the social world And the social world is a lot more complex and changing than the mountains
Given the complexity and changeability of the social world, we need to introduce some useful assumptions that make observing and measuring it different from observa-tion in the “natural” sciences First, we’re not going to say much about facts and knowledge in the strict sense
Chapter 1
Introduction
Learning Objectives
After studying this chapter, you should be able to:
1.1 Differentiate between qualitative and
quantitative methods in research.
1.2 Describe how the triangulation
methodology is used in research.
1.3 Analyze the general purpose of
Trang 13Quality refers to the what, how, when, where, and why of
a thing—its essence and ambience Qualitative research, thus, refers to the meanings, concepts, definitions, charac-teristics, metaphors, symbols, and descriptions of things
In contrast, quantitative research refers to counts and sures of things, the extents and distributions of our subject matter: how large a thing is, how many of them there are,
mea-or how likely we are to encounter one This distinction
is illustrated in Jackson’s (1968) description of classroom odors in an elementary school, data which defines a site
in terms that we would not want to quantify There are odors in our lives that recall specific places and times, just
as there are songs or colors that can do the same These memories evoke feelings based on their qualities, and not their quantities Qualitative research strategies provide perspectives that can prompt recall of these common or half-forgotten sights, sounds, and smells
The meanings that we give to events and things come from their qualities To understand our lives, we need qualitative research But can we really measure the unquantifiable essences of the phenomena that imbue our
lives? Can we ever, in a word, know? The answer is yes, though it is a qualified yes We can study and measure
qualities as collections of meanings, as a spectrum of states
of being, but not as precise and solid objects Qualities are like smoke; they are real and we can see them, but they won’t stand still for us or form straight lines for our rulers
to capture Clearly, qualitative research requires some cialized tools and techniques
spe-Qualitative and quantitative methods give us ent, complementary pictures of the things we observe Unfortunately, because qualitative research tends to assess the quality of things using words, images, and descrip-tions and most of quantitative research relies chiefly on computers, many people erroneously regard quantita-tive strategies as more scientific than those employed in qualitative research The error of thinking underlying this particular critique is that of confusing the study of imprecise subject matter with the imprecise study of subjects For this reason alone, qualitative researchers need to be more precise, more careful in their definitions and procedures, and clearer in their writing than most other scientists From my perspective, this means con-ducting and describing research that can stand the test
differ-of subsequent researchers examining the same enon through similar or different methods Qualitative research is a long hard road, with elusive data on one side and stringent requirements for analysis on the other Admittedly, this means that students have a lot to learn and not a lot of room for errors
phenom-What are these qualities that we measure? Why don’t
we quantify them? As for that second question, times we do, and sometimes we don’t All qualities can be quantified up to a point, just as all quantitative data have
some-We can make valid observations, measure real data, and
draw reliable and meaningful conclusions But to call
this knowledge “facts” might imply to some that they are
unchanging truths Everything we observe and measure is
only true up to a point So we talk about patterns,
tenden-cies, likelihoods, and generalities, but not facts
Second, though we are born into an existing
configu-ration of social, political, cultural, historical, and economic
circumstances, the social world is not simply out there
waiting to be found and understood It is socially
con-structed, continuously made and remade by human
activ-ity A single building, for example, can be understood as
an historical landmark, a tourist attraction, or an eyesore,
depending on whom you ask or when you ask that person
The building does not have to change for our
understand-ing of it to change There are fairly endurunderstand-ing social
struc-tures, ideas and practices that are deeply institutionalized
in our societies, and familiar tendencies among people
Still, all of those things are constantly open to challenge,
reconsideration, inertia, exaggeration, and other forms
of change Reality appears consistent, in part, because of
how we choose to define it So the observation of the social
world is necessarily an observation of choices and acts
made by people about the world
And third, as W I Thomas observed long ago, most of
the time we don’t need to worry about all that If we treat
the social world as though it’s just plain reality, it mostly
works It’s fairly stable and consistent because we believe
in it But it helps if our beliefs bear some resemblance to
empirical (measurable) reality And even if our partial
knowledge and impressionistic sense of things is enough
to get us through the day, much of it is still wrong
In the social sciences, we tend to favor quantitative
methods of data collection and analysis when we are seeking
to measure the relatively stable patterns and practices that
define our social structures; we adopt more qualitative
meth-ods when we need a deeper understanding of the exceptions
and special cases, or when we want to understand the
mean-ings and preferences that underlie those larger patterns
Quantitative work leans toward “what” questions, while
qualitative tends toward “why” and “how.” Like most
pat-terns of behavior, however, this distinction can be
mislead-ing until we really unpack how it works
1.1: Qualitative Methods,
Qualitative Data
1.1 Differentiate between qualitative and quantitative
methods in research
In his attempt to differentiate between quantitative and
qualitative approaches, Dabbs (1982, p 32) indicated that
the notion of quality is essential to the nature of things
Trang 14Introduction 13
people are politicians or celebrities Normatively, crime
is associated with violence and indirectly with poverty Similarly, sports coverage routinely incorporates athletic accomplishments, medical issues that threaten one’s abil-ity to play, and sports contracts But relatively little of it mentions endorsements, even though many athletes liter-ally wear their endorsements on their sleeves It seems that only some parts of the business of sports are widely perceived as related to sports Other aspects are placed in different categories We (as a society) come to recognize a certain cluster of things as belonging to the same category, and actively “split” other related things off into different categories, thereby creating “islands of meaning” out of the haphazard whirlwind of things in our lives (Zerubavel, 1996) We include 18-year-olds in our mental category for
“adults,” but not 17-year-olds These meanings might be codified into dictionary definitions that emphasize what is included But it takes more work to recognize those things that have been excluded
According to a study by Harold Garfinkel, one of the most immediate and effective ways to demonstrate the exis-tence of norms is to violate them and observe the results
A pattern of absences might or might not indicate that the exclusion of some class of events or people is considered normal But what happens when the usually excluded cat-egory is included?
Consider American movies Not only are most of the main characters straight, white, and presumably Christian men, but most of the random secondary characters seem to
be as well Women are introduced where the plot requires
a woman, as is true with nonwhites, gay characters, and others who are defined by their differences from the norm But is this evidence of norms at work, or just preferences and prejudices within a specific industry? One clue comes from those occasions when a film violates this expectation
by broadening the field of actors When a character is cast with a black actor (or defined as gay), is there pushback from viewers and critics? Is the casting decision derided
as “stunt” casting, even if the story does not require that the character be white (or straight)? If no ethnic or demo-graphic characteristics are required for the part, the popu-lar assumption is that the person will be whatever is most normative Thus, the expectations reveal the norms, and the objections to their violation, when they occur, reveal the expectations
Similar processes are at work in colleges, where fessors who include a diversity of materials are criticized
pro-by some students for this To add some sense of quantity
to this, professors who assign a majority of readings from white or male authors, with a small number of works by women or nonwhites, frequently report some number (a minority) of student evaluations accusing them of anti-male or antiwhite bias, as though the mere presence of any nonwhite expert or woman scholar is inherently suspect
qualitative aspects To better understand that, let’s
con-sider some of the qualities that we are good at measuring
One popular and important area of research
con-cerns social norms—the normatively expected and
infor-mally enforced patterns of behavior that are widely shared
within any given society Norms are vital to daily life in a
given society, as well as highly revealing about that society
But unlike rules, laws, and procedures, norms are almost
never written down or named This makes it a bit more
difficult to study them Nonetheless, they are visible to us
as researchers for exactly the same reasons that they are
visible to us as members of a culture We find evidence of
them everywhere
Jokes require and reveal norms Much of the work
of humor comes from surprising the listener by
violat-ing their expectations Jokes reveal both the normatively
expected and the normatively startling Racist, sexist, and
nationalistic jokes, for example, demonstrate the nature of
conventionally held negative ideas that one group of
peo-ple hold toward another In the United States in the 1960s,
for example, it was fairly conventional for newspapers
to print cartoons or jokes whose humor depended on the
stereotype that women were bad drivers But there were
probably no jokes at all about women as bad sign painters
Sign painting did not invoke or involve deeply held social
norms The driving jokes, however, reflected the
norma-tive assumption that most families had one car, that the
car belonged specifically to the man of the house, and that
his masculine prerogatives would have been threatened
by “allowing” his wife to drive At the same time, women
did drive and on average had better road safety records
than men So there was unarticulated social pressure to
continuously emphasize that driving was a naturally male
thing to do, hence the jokes, and men’s appreciation of
them Over time, as more middle-class families with two
adults became middle-class families with two jobs and
two cars, most people got used to the idea that American
masculinity was unharmed by sharing the road, and these
jokes became less popular (But they still show up once in
a while.) We use qualitative methods to interpret the jokes
and their underlying assumptions; we use quantitative
measures to show that they have fallen out of favor The
rise and fall of a style of joke reveals subtle shifts in social
norms over a period of a few decades
Absences also reveal norms Reviewing the content
of American newspapers, for example, demonstrates that
crime, politics, and entertainment are very important
ele-ments of what is considered newsworthy Yet, analyses
of the crime coverage show a preponderance of attention
to violent crime and “street” crime White-collar crime is
rarely mentioned at all, or only appears under the
head-ing of “business news.” It seems that the normative
per-ception of crime does not include the kinds of economic
crimes committed by people with money, unless those
www.downloadslide.com
Trang 15write, or at least endorse, their own words, and that they are important Analysis of news articles in the study of key social events relies on the assumption that key events are represented with descriptive accuracy in the news Each method, thus, reveals slightly different facets of the same symbolic reality Every method is a different line of sight directed toward the same point, observing particular aspects of the social and symbolic reality By combining several lines of sight, researchers obtain a better, more substantive picture of reality; a richer, more complete array
of symbols and theoretical concepts; and a means of ing many of these elements The use of multiple lines of
verify-sight is frequently called triangulation.
“Triangulation” is a term originally more common in surveying activities, map making, navigation, and military practices In each case, three known points or objects are used to draw sighting lines toward an unknown point or object Usually, these three sighting lines intersect, forming
a small triangle called the triangle of error The best estimate
of the true location of the new point or object is the center
of the triangle, assuming that the three lines are about equal in error Although sightings could be done with two sighting lines intersecting at one point, the third line permits a more accurate estimate of the unknown point or object (Berg & Berg, 1993)
Triangulation was first used in the social sciences as
a metaphor describing a form of multiple operationalism or convergent validation (Campbell, 1956; Campbell & Fiske,
1959) In those cases, triangulation was used largely to describe multiple data-collection technologies designed to measure a single concept or construct (data triangulation) However, Denzin (1978, p 292) introduced an additional
metaphor, lines of action, which characterizes the use of
multiple data-collection technologies, multiple theories, multiple researchers, multiple methodologies, or combi-nations of these four categories of research activities (see Figure 1.1)
For many researchers, triangulation is restricted to the use of multiple data-gathering techniques (usually three) to investigate the same phenomenon This is inter-preted as a means of mutual confirmation of measures and validation of findings (Casey & Murphy, 2009; Leedy, 2001; Leedy & Ormrod, 2004) Fielding and Fielding (1986,
p 31) specifically addressed this aspect of triangulation They suggested that the important feature of triangulation
is not the simple combination of different kinds of data but the attempt to relate them so as to counteract the threats to validity identified in each
Denzin insists that the multiple-methods approach is the generic form of this approach But triangulation actu-ally represents varieties of data, investigators, theories, and methods Denzin (1978, p 295) outlined these four categories into more detailed subgroupings of time and place, social setting, theoretical perspective, and mixed
Now it is important to note that usually the majority of
students don’t complain, the professors are not punished,
and the classes continue to run No free speech rights are
on the line The point is not that the faculty is prevented
from teaching the work of black authors or anyone else
The point is that some members of the dominant culture
think that such a thing as diversity is odd The fact that
they would make an issue of it demonstrates the presence
of the social norms; their complaints reveal what they
expected to find
In each of these cases, I am describing how the
exis-tence of specific social norms may be demonstrated
through the qualitative analysis of what we call social
arti-facts—things produced or performed by people in the
nor-mal course of their lives Two very important points need
to be emphasized about these examples First, I am not
describing a single event as evidence of social values, but
rather a regular and familiar pattern of events Individual
cases may not mean very much We tend to look instead
at multitudes of cases And second, these cases reveal
the existence of specific norms, and not the number of
people who adhere to them, the strength of people’s belief
in them, or the likelihood of encountering them That is,
we can’t quantify this data based on the kinds of studies
described here That sort of question requires different
Most researchers have at least one methodological
tech-nique they feel most comfortable using, which often
becomes their favorite or only approach to research
Furthermore, many researchers perceive their research
method as an atheoretical tool, distinct from the conceptual
frameworks that shape their research questions (Denzin,
1978) Because of this, they fail to recognize that methods
impose certain perspectives on reality For example, when
researchers canvass a neighborhood and arrange
inter-views with residents to discuss some social problem, a
the-oretical assumption has already been made—specifically,
that reality is fairly constant and stable and that people
can reliably observe and describe it Similarly, when they
make direct observations of events, researchers assume
these events are deeply affected by the actions of all
par-ticipants, including themselves (I’m not saying that this is
not a fair assumption, only that it is a more or less hidden
assumption that precedes the application of “theory.”)
Content analysis of important speeches generally relies on
the assumption that the people who give these speeches
Trang 16Introduction 15
techniques in addition to multiple data-collection cedures The use of multiple research design strategies and theories increases the depth of understanding an investigation can yield (see also Dittmann, 2005; Miles & Huberman, 2002)
pro-1.3: Qualitative Strategies: Defining an Orientation
1.3 Analyze the general purpose of qualitative data
We do not conduct research only to amass data The purpose
of research is to discover answers to questions through the application of systematic procedures Qualitative research properly seeks answers by examining various social settings and the groups or individuals who inhabit these settings Qualitative researchers, then, are most interested in how humans arrange themselves and their settings and how inhabitants of these settings make sense of their surround-ings through symbols, rituals, social structures, social roles, and so forth
Research on human beings affects how these persons will be viewed (Bogdan & Taylor, 1998) When humans are studied in a symbolically reduced, statistically aggregated fashion, there is a danger that conclusions—although arith-metically precise—may misrepresent the people or circum-stances studied (Mills, 1959) Qualitative procedures seek patterns among cases, but do not reduce these cases to their averages They provide a means of accessing unquan-tifiable knowledge about the actual people researchers observe and talk to or about people represented by their personal traces (such as letters, photographs, newspaper
methods It is difficult for a single text or course to prepare
students to accomplish all that Triangulation, as a model
for research, requires researchers to be fluent in multiple
methods Yet, it is useful to study qualitative and
quantita-tive techniques somewhat independently, if only to give
each its due credit
Some authors of general-purpose research texts
associ-ate qualitative research with the single technique of
partici-pant observation Other writers extend their understanding
of qualitative research to include interviewing as well
However, qualitative research also includes such
meth-ods as observation of experimental natural settings,
pho-tographic techniques (including videotaping), historical
analysis (historiography), document and textual analysis,
sociometry, sociodrama, and similar ethnomethodological
experimentation, ethnographic research, and a number of
unobtrusive techniques In the interests of triangulation,
primarily qualitative studies need not exclude quantitative
data-gathering techniques as well, though we won’t be
dis-cussing them here
This book stresses several discrete yet intertwined
strategies and techniques involved in each of the major
research schemes In fact, the decision to discuss field
research strategies under the broad umbrella of ethnography
ensures the inclusion of a wide combination of elements,
such as direct observation, various types of
interview-ing (informal, formal, semiformal), listeninterview-ing, document
analysis (e.g., letters or newspaper clippings), and
ethno-methodological experimentation Novice researchers are
thus instructed in the use of research strategies composed
of multiple methods in a single investigation I also
fol-low Denzin’s (2010) approach that triangulation includes
multiple theoretical perspectives and multiple analysis
Multiple Theories
Multiple Researchers
Multiple Data Technologies
Multiple Lines
of Action
Research Findings
Research Idea
Figure 1.1 Multiple Lines of Action in Triangulation
www.downloadslide.com
Trang 17a pattern, not a law Exceptions neither prove nor disprove the tendency.
It’s been my observation that people don’t like plete information, or generalizations, that can’t be applied universally We should test that idea before making too many assertions, but I believe this to be fair I think it’s one of the reasons that people both oversimplify social reality and think that research oversimplifies This leads to
incom-what I like to call the life cycle of a sociological study It works
something like this:
1 A researcher notices an interesting thing and decides
to look into it For example, it might be that pet ers who have daily conversations with their parrots claim that this is great for reducing stress, and we want to know more about this idea
own-2 The researcher adopts a set of stress measures ably quantitative) and a measure of the quality of one’s relationship with pets She designs a study for some number of participants across the spectrum of pet ownership, gets funding and approvals, and be-gins to collect data
(prob-3 The results indicate that people who have “good” tionships with their pets are less stressed than people who don’t (I’m making this example up; no promises for you pet owners.)
rela-4 The researcher writes a paper in which she discusses all of the major issues around stress and stress relief, including past research with animals, the health risks
of high stress, and the problems of social isolation She concludes that talking with your animal companions, particularly birds, can be part of a healthy lifestyle, qualifying this to remind readers that it could well be that people who are mellow enough to talk to their birds might not have been all that stressed to begin with
5 The paper’s publisher distributes the abstract, which states that conversations with pets are associated with low-stress, heart-healthy lives
6 Some news or entertainment media source picks this
up and broadcasts, “Can Talking to a Bird Save Your Life??!” In their full story, they speak with “lifestyle” experts, some of whom say it makes sense, and some who say it probably doesn’t None of them discuss any
of the methods, qualifications, or limitations actually described in the article, simplifying the whole thing to either “science says that you should talk to your bird”
or “this one scientist thinks that raising birds is more important than exercise.”
7 Scores of people write comments to the news sites, ing things like “what is wrong with those sociologists who keep claiming to be saving the world with their trivial studies?” or “We all knew that already! What a waste of money.” And inevitably, “This is stupid I know someone who took care of 10 birds and still died.”
say-accounts, and diaries) As a result, qualitative techniques
allow researchers to share in the understandings and
per-ceptions of others and to explore how people structure and
give meaning to their daily lives Researchers using
quali-tative techniques examine how people learn about and
make sense of themselves and others Of course, the more
depth of knowledge you have of a particular group, the
more you capture the uniqueness of that group An
advan-tage that much quantitative research has over qualitative is
that it ignores this unique depth in favor of a more general,
widespread pattern of acts or ideas In other words,
quali-tative research does not generalize as easily over a large
population
Before we get too much into the nature of the
limita-tions on our data, we need to be clearer about the uses
of this data I said earlier that we deal in patterns, not
facts What does that mean? Let us suppose that we have
conducted a series of interviews with Chicago Cubs fans
and found that a large number of them appear to have
adopted a sense of fatalism about their team’s prospects
That is an interesting finding in itself, but to claim that a
“large number” of them have this quality does not mean
that they all share this quality, or that this quality is caused
by rooting for the Cubs, or that they approach everything
in their lives this way It does mean that there is a pattern
among the responses from the fans that stands out as
dif-ferent from what is known about the general population
This pattern can gain some explanatory power when we
compare the respondents’ feelings about the team with
their feelings about other aspects of their lives It can tell
us something about baseball fans if we were to compare
this group with Yankees fans or Royals fans And we can
certainly make meaningful comparisons with other groups
of people who have waited a very long time for something
they wish would happen, but maybe no longer believe in
Presenting this data can be very tricky If I say that
a great many fans of this team share a certain attitude
toward the team, I need to be careful not to
overgeneral-ize and imply that you have to have this attitude to follow
the Cubs Obviously, it would not be hard to find one fan,
among that many, who completely contradicts this idea
What’s important to know is that this exception, or many
exceptions, doesn’t matter, because we are not trying to
make a big claim about everyone The patterns we find
are real and have significance even though they are not
absolute rules that need to apply to all people The same is
true for the other examples discussed already If a review of
popular contemporary movies finds, as suggested above,
that the generic assumption for all major characters is that
they are white heterosexual men, then this shows that some
sort of filtering process is happening in the film industry at
some level, whether it’s in the writing, the directing, or just
the casting And this finding remains meaningful
regard-less of how many starring roles Morgan Freeman has It’s
Trang 18Introduction 17
correct—if men define situations as real, they are real in their consequences” (Thomas & Swaine, 1928, p 572).For instance, the first day of each semester, students walk into their classroom and see someone who appears
to be the professor This supposed professor begins to ture, distribute syllabi, discuss course requirements, and conduct various other traditional first-day activities Few,
lec-if any, students ask to see their professor’s credentials Yet, the students, within certain limits, perform their roles as students so long as this professor continues to perform the role of instructor Suppose that several weeks into the semester, however, the class is notified that the person they assumed to be a professor is really a local dogcatcher who has no academic credentials The question then becomes whether the reality of the classroom experi-ence during the previous weeks is void merely because the dogcatcher was incorrectly interpreted as a profes-sor It would, of course, remain to be seen whether any information conveyed by the dogcatcher was accurate, and certainly, the classroom remained a classroom and students continued to perform their expected roles From Thomas’s perspective, these youths had defined the real-ity as a class, and it became one for them Interestingly,
a real version of this scenario confronted the University
of Chicago and its students when it was revealed that the celebrated psychology professor Bruno Bettelheim, who had taught there for 30 years until the early 1970s (the same decades during which Chicago sociologists were developing the theory of symbolic interactionism), had faked his academic credentials and was not actually trained as a psychologist What, then, are we to make of his research findings or his teachings? Have they ceased
to qualify as knowledge?
Symbolic interactionists tend to differ slightly among themselves regarding the relative significance of various aspects of an interactionist perspective Several basic ele-ments, however, tend to bind together even the most diverse symbolic interactionists First, all interactionists agree that human interactions form the central source
of data Second, there is a general consensus that ticipants’ perspectives and their ability to take the roles
par-of others (empathy) are key issues in any formulation par-of
a theory of symbolic interaction Third, interactionists agree with Thomas concerning “definitions of a situation,” that is, the view that how inhabitants of a setting define their situation determines the nature and meaning of their actions as well as the setting itself
Researchers in different schools have given rise to different schools of thought within the interactionist par-adigm While the idea originated at the University of Chicago under the leadership of Herbert Blumer, Manford Kuhn and researchers at the University of Iowa developed their own approach Among the more prominent contribu-tions to symbolic interaction from the Iowa School is the
In simpler terms, we design and conduct careful,
qual-ified research that indicates partial relationships among
important social variables and which sets these
relation-ships in a context Other people, looking for permanent
social laws, tear these results out of context and claim
too much for them Then the researcher is blamed for the
excesses
The moral, though, is to do careful work, note its
limi-tations, and try not to be quoted out of context We
coun-teract misunderstanding and misinformation with clarity
and caution
This explanation of the general purpose of qualitative
research in which we are searching for interpretive
pat-terns of meaning derives from a symbolic interactionist
perspective Symbolic interaction is an umbrella concept
under which a variety of related theoretical orientations
may be placed The theme that unites the diverse
ele-ments of symbolic interaction is the focus on subjective
understandings and the perceptions of and about people,
symbols, and objects
1.4: From a Symbolic
Interactionist Perspective
1.4 Examine symbolic interactionism as a school of
thought of the social sciences
Symbolic interactionism is one of the several theoretical
schools of thought in the social sciences The substantive
basis for symbolic interaction as a theory is frequently
attributed to the social behavioral work of Dewey (1930),
Cooley (1902), Parks (1915), Mead (1934, 1938), and several
other early theorists, but Herbert Blumer is considered the
founder of symbolic interactionism In fact, he coined the
term In articulating his view of what symbolic
interac-tion is, Blumer (1969) first established that human beings
account for meaning in two basic ways First, meaning may
be seen as intrinsically attached to an object, event,
phe-nomenon, and so on Second, meaning may be understood
as a “psychical accretion” imposed on objects, events,
and the like by people As Blumer (1969, p 5) explained,
“Symbolic interactionism sees meanings as social products
formed through activities of people interacting.” Objects
and events exist Meaning is attached to them by human
thought and action
Blumer thereby suggests that meanings derive from
the social process of people or groups of people interacting
Meanings allow people to produce various realities that
constitute the sensory world (the so-called real world),
but because these realities are related to how people
cre-ate meanings, reality becomes an interpretation of various
definitional options Consequently, as referenced earlier,
“It is not important whether or not the interpretation is
www.downloadslide.com
Trang 19considerable measure, this involves the issue of ity In other words, when one considers deterministically what causes certain events, this understanding bears on
causal-the methodology used From Blumer’s (1969) istic orientation, social structures are to be understood
indetermin-as emergent phenomenon, and, in effect, indetermin-as the product
of shared interpretations held by people Consequently, these understandings are the result of internal symbolic processes that allow an individual to group together vari-ous behaviors into an organized coherent pattern, such that it offers meaning These understandings, however, are not the result of system forces, societal needs, or struc-tural mechanisms Social organization from this point of view is the result of mutual interpretations, evaluations, definitions, and social mappings created by individuals (Herman, 1995) For Blumer and his followers, the sym-bolic processes of humans cannot be conceived as a mech-anism through which social forces operate; rather, they must be viewed as shaping the way structures are created, maintained, and transformed In this sort of orientation, it
is difficult to establish causality Social structures or
orga-nizations do not cause human behaviors; instead, these are
development of a research instrument called the
twenty-statement test (TST) The TST can be used to identify
self-designations that result from social roles an individual
plays rather than from his or her personal self-concepts
The TST is a rather simple tool that asks the subject the
question, “Who am I?” The subject then fills out 20 blank
spaces in answer to this question The responses are scored
as representing either an external or internal self-concept
Figure 1.2 offers an example of the TST
The twenty-statement test can be used for a rough
assessment of an individual’s sense of self or identity The
test has the virtues of being straightforward and simple
and providing a relatively direct measure of the subject’s
self-concepts In contrast to this systematic orientation, the
Chicago School’s orientation relied more heavily on
partici-pant observational research Thus, the Chicago School was
somewhat more anthropological and sought to understand
the meanings of individuals and groups without an
empha-sis on revealing generalizable patterns of human behavior
The differences between Blumer’s and Kuhn’s
meth-odological approaches center on their assumptions
concerning the operation of symbolic processes To a
Figure 1.2 The Twenty-Statement Test
Please write 20 answers to the question “Who am I?”
Scoring Instructions: Categorize each of the twenty statements in terms of each
giving a description of the subject as external or internal.
External: This phrase locates the individual in society by describing some social role
he or she plays or enacts For example, the names of social roles one holds are all
external: mother, father, son, daughter, student, salesman, police officer, store clerk,
baseball fan, and so on.
Internal: This phrase locates the individual inside his or her self by describing an internal
or interior quality or trait one possesses For example, names of personal intrinsic
qualities or characteristics one possesses are all internal: shy, ambitious, insecure,
happy, sad, ambiguous, curious, depressed, hard working, industrious, and so on.
Place an E for external or an I for internal beside each of the 20 statements; then, total
up the number of statements representing each category.
Total number of external descriptors:
Total number of internal descriptors:
Trang 20Introduction 19
source of entertainment and pleasure, while for the inmate held in a maximum security prison who watches home movies sent from his or her family, it may be considered
a window to the outside world The meanings that people attach to their experiences and the objects and events that make up these experiences are not accidental or uncon-nected Both the experiences and the events surrounding them are essential to the construction of meanings One could view the DVD player as a single, unambiguous device with many possible uses But to do so implies that objects and events have an inherent reality distinct from their meanings The interactionist perspective assumes that the key to defining an object or event is found in the meaning that users attribute to it
To understand behavior, one must first understand the definitions and meanings and the processes by which they have been created Human behavior does not occur on the basis of predetermined lockstep responses to preset events
or situations Rather, human behavior is an ongoing and negotiated interpretation of objects, events, and situations (Bogdan & Biklen, 2003) For researchers to understand the meanings that emerge from these interactions, they must either enter into the defining process or develop a sufficient appreciation for the process so that understand-ings can become clear A disk player is just a device and may be deceptively simple to analyze But what of a some-what chaotic, somewhat disruptive protest march Does it have an underlying reality, separate from meaning? Is it a demonstration of solidarity, an attack on social order, an exemplar of democracy in action, or a coordinated crimi-nal event? Can the event be explained without adopting a perspective and system of meaning?
Although social roles, institutional structures, rules, norms, goals, and the like may provide the raw material with which individuals create their definitions, these elements do not by themselves determine what the definitions will be or how individuals will act In essence, symbolic interactionism emphasizes social interactions (action with symbolic mean-ing), negotiation of definitions, and emphatic role-taking between humans (Gecas, 1981; Turner, 1978) Measuring these interactions forms the core of the data-collection strate-gies that we will be studying in this book
1.5: Why Use Qualitative Methods?
1.5 Recognize the significance of the right tools for effective qualitative research
It has been suggested that to a child with a hammer, every problem looks like a nail We all have our preferred meth-ods of dealing with challenges Specialization can be quite useful, but it has its downside as well
merely types of objects in the individual’s environment
and symbolic thought processes Research, therefore, must
focus on subjects’ meanings, expectations, and perceptions
first, with actions and decisions following
In contrast, Kuhn argued for a deterministic model of
social organization From this perspective, social
institu-tions are viewed as representing relatively stable networks
of social positions accompanied by associated norms and
expectations Symbolic interactions between individuals,
then, are adept at creating and altering situations and
structures Once these structures are created, they are
capable of constraining individuals From this perspective,
social structures are understood as fairly stable, especially
when the individual’s core self is invested in these social
structures and networks of positions If one can learn
about the nature of one’s core self, of the expectations one
has internalized, as well as one’s expectations in a given
situation, it is possible, according to Kuhn, to predict
peo-ple’s definitions of a situation, as well as their behaviors
The social setting constrains much of the meaning systems
that the people in the setting use, and this setting can be
studied independently of the people in it
These divergent assumptions about human
behav-ior and issues of causality resulted in followers of the
Chicago School and the Iowa School adopting different
methodological approaches That is to say, different
theo-ries shaped different research methods Blumer and his
followers borrowed from the phenomenologists and
ori-ented their methodological strategies toward
nongeneral-izing and idiographic methods The primary goal of this
approach was to make social life intelligible From this
perspective, the act of research must be viewed as a
pro-cess of symbolic interaction wherein the researcher takes
the role of the subjects who are being studied Blumer and
his followers, then, saw research as possessing a twofold
agenda: (1) exploration, where the researcher examines
and observes specific situations and events, followed by
(2) inspection wherein the researcher uses data
(systemati-cally collected) to refine concepts, and then to use these in
general statements describing human life and behavior
In contrast to this, Kuhn and his followers maintained
a deterministic emphasis, stressed the commonality of
methods across all the sciences, and tended to follow the
basic principles of logical positivism From this
perspec-tive, the goal of methodology is to specify operational
definitions of concepts that can be tested (Herman, 1995;
Maines, Sugrue, & Katovich, 1983) Objects, people,
situ-ations, and events do not in themselves possess meaning
Meaning is conferred on these elements by and through
human interaction For example, a DVD player in a college
classroom may be defined by the professor as a teaching
device to be used for showing educational videos For
the student using a DVD player in his or her dormitory
to view rented movies, this instrument may be seen as a
www.downloadslide.com
Trang 21much-needed assistance for all researchers, including the inexperienced, through a discussion of various qualitative research strategies, design development, data organization and presentation, and analysis procedures.
We now offer the ninth edition of this book, once again focusing on innovative ways of collecting and analyzing qualitative data collected in natural settings I continue to address those data-collection strategies that may be char-acterized as the building blocks for emerging research-ers As in past editions, this text concentrates on basic procedures This text avoids the cookbook approach to research; very few instruction lists or absolute statements
of what you must do for your research fully represent one technique or other Instead, my goal is to offer a handle on what these techniques are; why, when, and how we use them; and what we can get out of them Of course, this also includes cautionary notes about their limitations and a cer-tain amount of attention to when not to use each approach Throughout, I make a few simple assumptions First, if you are reading this book, it means you are training to do research and, therefore, probably want to know how to take charge of your own projects and get the good results that will answer your questions Second, if you want to apply some specific technique or creative combination of techniques, but want more of a checklist to go with it, you know how to find one I’m not saying that such things aren’t useful, only that my priorities lean more toward depth of understanding and away from vocabulary tests and recipes Finally, I assume that the first draft of any-thing any of us comes up with will not be sufficient For that reason, I imagine that you, students, will be reading parts of the chapters for instructions on how to get started, and then returning for ideas about how to fix whatever design or plan you have started on The organization of most chapters is intended to support such an approach This new edition continues the impossible task of try-ing to keep up with developing technologies, incorporates recent examples of important and innovative qualitative research, and strengthens the presentation of basic tech-niques As well, this edition goes further in attempting to integrate all of the material into a cohesive lesson on plan-ning and carrying out your research, with more explana-tion of research design and more attention to design issues throughout the chapters I also provide new material on the very important questions of when not to use certain techniques and when and how specific techniques can fail
to serve
This book describes in detail seven primary ways to collect qualitative data: interviewing, focus groups, eth-nography, observations, historiography, content analysis, and case studies In addition, we will examine a framework for undertaking participatory research studies, sometimes called action research Action research has a substantial
Many researchers believe that the social sciences have
depended too much on sterile survey techniques,
regard-less of whether the technology is appropriate for the
prob-lem For instance, nurses, when taught to do research at all,
are strongly urged to use scientific strategies of
quantifica-tion over more sociologically or anthropologically oriented
ones that are considered less scientific Unfortunately,
clinical settings in which nurses are likely to conduct their
research fail to meet most quantitative requirements for
representativeness and sufficiency of sample size to allow
statistically meaningful results The tools at their disposal
are not the right ones for the job
For instance, let us say the average number of beds
in a critical care unit varies between 8 and 12 Even when
there are multiple units (e.g., in a medical intensive care
unit or a cardiac intensive care unit), typically, fewer than
40 cases are available at any given time With regard to
research strategy, such a situation should preclude most
quantitative investigations On the other hand, 40 cases
would prove ample for a number of qualitative strategies
In fact, as Chapter 8 describes, a setting such as a hospital
would provide researchers with numerous opportunities
to implement unobtrusive measures
We believe that researchers need a complete tool kit
with which to craft the best approach to any given problem
or topic Scientific researchers may emphasize a more
posi-tivist view or may be primarily interested in individuals
and their so-called life-worlds In the case of life-worlds,
researchers focus on naturally emerging languages and
the meanings individuals assign to experience Life-worlds
include emotions, motivations, symbols and their
mean-ings, empathy, and other subjective aspects associated
with naturally evolving lives of individuals and groups
These elements may also represent their behavioral
rou-tines, experiences, and various conditions affecting these
usual routines or natural settings Many of these elements
are directly observable and as such may be viewed as
objectively measurable data Nonetheless, certain elements
of symbolism, meaning, or understanding usually require
consideration of the individual’s own perceptions and
subjective apprehensions This is qualitative data
1.6: A Plan of Presentation
1.6 Report how the book helps students of the social
sciences
Colleges require students to study research methods both
to learn the major work of our fields of study and to acquire
pragmatic skills Thus, students must confront the myriad
problems associated with understanding empirical results,
as well as the process of research itself This book provides
Trang 22Introduction 21
into the natural setting by examining ethnography Along with interviewing, Chapter 6 discusses watching and lis-tening, field notes, and a number of other field research concerns This chapter examines ethnography both as
a means of collecting data (what some call the new nography) and as an end in itself (narrative ethnographic
eth-accounts) This chapter further explores critical raphy and the role it may play in the ethical conduct of naturalistic research
ethnog-Chapter 7 considers a dynamic mode of research, namely, action research Action research has a substantial history in educational and nursing research and is moving rapidly into broader scientific endeavors as well
While Chapters 4, 5, and 6 separately address the concept of interviewer reactivity, Chapter 8 offers several strategies that avoid reactivity almost entirely: It explores the use of unobtrusive measures
As foreshadowed slightly in Chapter 8, the use of certain unobtrusive data has grown quite specialized Chapter 9 examines a specialized and systematic use of certain kinds of running records, namely, historiography
In addition to the use of records, Chapter 9 considers oral histories and life histories as variations in historiography.Chapter 10 examines a technique used to study individuals in their unique settings or situations This
technique is commonly called the case study method This
chapter also discusses how case studies may be taken on communities and organizations
under-Chapter 11 dovetails with each of the preceding ters on research techniques Included in this chapter are recommendations for how novice researchers may orga-nize their data and begin to make sense of what may be volumes of notes, transcripts, and trace documents and artifacts Chapter 11 also briefly discusses the use of com-puters to assist in this data management scheme
chap-Chapter 12, the final chapter, offers recommendations for how novice qualitative researchers can disseminate their research findings
“Trying It Out,” a section at the conclusion of each of the data-collection technique chapters, offers suggestions for practicing each of the seven strategies Most chapters also contain a “Why It Works” section and a “Why It Fails” section highlighting conditions that are or are not compat-ible with the technique under discussion
history in educational and nursing research and is moving
rapidly into broader scientific endeavors as well These
methods include an examination of the basic theoretical
assumptions of each technique and advice on how to start
each procedure and how to resolve problems that may
arise Furthermore, I present the technique of content
anal-ysis as the model for the analanal-ysis of most qualitative data,
particularly those that we call “social artifacts.” Also as an
essential element or consideration in any research study,
this book explores the ethical dimensions of conducting
research on humans; it is within the context of this ethical
dimension to research that the section on critical
ethnogra-phy has been included This edition of Qualitative Research
Methods for the Social Sciences begins with the assumption
that the reader knows little or nothing about the research
process Chapter 2, therefore, offers a basic description of
how to design a research project Most of the rest of the
book can be read in almost any order
Having briefly outlined the basic assumptions and
qual-itative orientations of symbolic interaction, it is now possible
to weave in various methodological strategies Chapter 2
provides the basic information necessary for understanding
the research enterprise This chapter discusses the research
process and proposes a spiraling model to follow when
developing a research agenda Chapter 2 also offers advice
about how to organize and conduct a literature review
Chapter 3 considers a number of ethical concerns that
are important for new investigators to understand before
actually conducting research Among the salient issues
considered are covert versus overt research concerns,
pri-vacy rights, human subject institutional review boards,
and informed consent in human subject research
In addition to providing a general discussion of
vari-ous forms and styles of traditional interviewing
tech-niques, Chapter 4 uses a kind of symbolic interaction
known as dramaturgy and suggests an effective research
strategy for conducting in-depth interviews
Chapter 5 also addresses the area of interviewing but
moves toward a specialized style, namely, focus groups
This chapter examines the early origins of focus group
interviews, their development during the past several
decades, and their growing use in the social sciences
Chapter 6 builds on the foundation constructed in
Chapters 1 through 4 and extends the research process
www.downloadslide.com
Trang 23This chapter considers various ways of thinking about and
planning research If you don’t know where you’re going,
George Harrison observed, any road will take you there
But if you do have a particular destination in mind, then
it’s pretty important to choose your path deliberately and
carefully In research terms, we have a lot of tools and
tech-niques that are discussed in this book, but you have to
de-cide which you need when, and why, and how to apply it
to your research problem
This chapter will get you started on planning your
research journey It includes discussion of the
relation-ships among ideas, theory, and concepts and of what
many people find to be the most difficult facet of research:
conceptualization This chapter further offers a strategy
for conducting literature reviews and explains the
impor-tance of carefully designing and planning research in
advance Let’s begin with some thoughts about ideas,
concepts, and theory
2.1: Theory and Concepts
2.1 Evaluate the applicability of theory and concepts
in qualitative research
In the natural sciences, certain patterns of relationships occur with such regularity that they are deemed laws: occurrences of universal certainty No such laws are found
in the social sciences This does not, however, mean that social life operates in a totally chaotic or completely irra-tional manner Rather, social life operates within fairly regular patterns, and when carefully examined, these pat-terns make considerable sense Unlike laws, patterns are tendencies, representing typical and expected forms of action around which innumerable individual variations may be found As well, patterns of expected action often include smaller patterns of reaction against the expected actions It is as though for every large group of balls that
Chapter 2
Designing Qualitative Research
Learning Objectives
After studying this chapter, you should be able to:
2.1 Evaluate the applicability of theory and
concepts in qualitative research.
2.2 Explain how research progresses from the
original idea.
2.3 Describe the importance of authentic
literature in research.
2.4 Give an example of a problem statement
with researchable questions.
2.5 Describe the process of operationally
defining a concept.
2.6 Examine how the technique of concept
mapping assists the research design process.
2.7 Recognize the importance of advance planning before beginning the data- collection process.
2.8 Describe the three concurrent flows of action comprising data analysis.
2.9 Explain why dissemination of research findings is important.
2.10 Analyze why the design logic is important
in understanding research.
2.11 Recognize why research fails at times.
Trang 24Designing Qualitative Research 23
of the world, a large part of what we are referring to is the process of grouping some forms of behavior under one name, others under a different name, and not naming some at all These groups are named in order to convey some concept For example, different societies conceptualize “family” differ-ently, and each will have in mind a somewhat different set of relations when they use that word Similarly, many societies divide the world of animals into such groupings as “pets,”
“food,” “work animals,” and “wild.” We treat these divisions
as though they are simply elements of the natural world and not reflections of our own social relations with nature These groupings vary and are almost arbitrary Yet, when one cul-ture sees an animal as a pet and another sees it as food, mem-bers of each culture are likely to feel that their own definitions are simply true and that the others are weird Conceptual definitions of things reflect how we choose to understand the things that we are defining
In terms of ideas, concepts are important because they are the foundation of communication and thought Concepts provide a means for people to let others know what they are thinking and allow information to be shared Thus, instead of describing a youth who is involved with drugs, crime, or tru-ancy, or has problems with parents and other adults, I might
simply use the concept of delinquent to communicate these same elements (ideas) By conceptualizing a set of behaviors or
ideas as part of a coherent package, we can describe a range
of possible ideas, relations, and outcomes with a single term Since concepts are abstract representations; of course, they contain a much broader range of possibilities than what any individual case is likely to contain Most delinquent youths, for example, are not all that delinquent, while others are so far out there that we might prefer the term “criminal.”
Concepts can be found everywhere, and people use them all of the time without actually thinking about them as concepts (Silverman, 2006) For example, age is
a concept that is so commonly used that few people stop
to think about what it means Even though people often think they understand the meaning of the concept, they may hesitate when asked to offer a specific definition We often use precise numbers to describe ages when we are really seeking to communicate abstract concepts, such as
“young” or “elderly.” Or we mentally translate such terms from the abstract “middle-aged” to some approximate age range All of this is dependent of context as well A jazz musician might seem fairly young at the age of 50, while a football player is getting old at 29
As data, age actually represents an abstract idea about the number of cumulative years that an individual has been alive In research, other related ideas, such as health
or infirmity, stage in the life course, or work experience, must be specified separately rather than assumed as attri-butes of one’s age Although this may seem to make the term stiff, it also ensures that there is a common under-standing for the meaning of this concept Concepts used
fall down, a few fall up or to the side Gravity defines the
general pattern, while other actions unrelated to gravity
form a smaller pattern within the whole
One purpose of social scientific research is to find
the meaning underlying these various patterns This is
accomplished by creating, examining, testing, and refining
theory What then is theory? Theory is the meaning that
we assign to things that we observe in order to make sense
of them Theory can be defined as a general and more or
less comprehensive set of statements or propositions that
describe different aspects of some phenomenon (Hagan,
2006; Silverman, 2006) In an applied context, theories can
be understood as interrelated ideas about various patterns,
concepts, processes, relationships, or events In a formal
sense, social scientists usually define theory as a system of
logical statements or propositions that explain the
relation-ship between two or more objects, concepts, phenomena, or
characteristics of humans—what are sometimes called
vari-ables (Babbie, 2007; Denzin, 1978; Polit, Beck, & Hungler,
2003) Theory might also represent attempts to develop
coherent narratives about reality or ways to classify and
organize events, describe events, or even predict future
events (Hagan, 2006) Theories are explanations The theory
of gravity explains why things fall, as well as predicting and
explaining orbits and the physical stability of the universe
Theories of inequality contribute to our explanations for all
kinds of economic behavior, from consumption to crime
to wedding receptions In time, we may find newer and
more informative ways to explain the things we experience
as gravity, or the ways in which we respond to inequality
These new approaches may take on different names, but
that will not mean that the original theories were wrong,
only that explanations can be improved with more data
Theories have general applicability I would not, for
example, theorize that the shelf above my bathroom sink will
collapse if I put more stuff on it I would theorize that certain
construction materials have limited weight capacity, which
can be exceeded I might theorize that when there are more
objects to be shelved than there are shelves to hold them,
people will frequently choose the short-term convenience
of placing too many things on one shelf over the long-term
benefit of building or finding new places to put things These
two theoretical models together yield a tangible prediction: I
have to do something about all of this junk or my shelf will
fall That last prediction is more of a hypothesis—a testable
proposition about specific cases or variables
In order to construct theories, one needs some smaller
components or what Jonathan Turner (1989, p 5) calls the
“basic building blocks of theory,” namely, concepts Concepts,
then, are symbolic or abstract elements representing objects,
properties, or features of objects, processes, or phenomenon
Concepts may communicate ideas or introduce particular
perspectives, or they may be a means for explaining a broad
generalization When we talk about the social construction
www.downloadslide.com
Trang 25Propositions, then, are statements about relationships between concepts (Maxfield & Babbie, 2007) Taylor and Bogdan (1998) suggest that although a concept may fit or not (may or may not convey the intended meaning), propositions aim to be either right or wrong statements of fact, although the research may not be able to prove them Testable propositions about the relations among our research concepts form a special class
of propositions called hypotheses Propositions, as discussed
later, are the statements that make up theories
2.2: Ideas and Theory
2.2 Explain how research progresses from the original idea
Every research project has to start somewhere; typically, the starting point is an idea The big question, however, is how
to go about finding an idea that will serve as a good ing point to a research project For some students, this genuinely is the most difficult part of the research process Actually, many people arrive at their research ideas sim-ply by taking stock of themselves and looking around For example, a nurse might observe a coworker coming to work under the influence of alcohol and begin to think about how alcohol would influence nursing care From this initial thought, the idea for researching impaired nurses might arise A counselor at a delinquency detention center might notice that many of her clients have been battered or abused prior to their run-in with the law From her observation, she might wonder how abuse might be linked with delinquency and how she could investigate this linkage
launch-In some situations, ideas derive from information you hear but may not actually experience yourself For instance, you’re sitting at home listening to the news, and you hear a report about three people from wealthy families who have been caught burglarizing houses You wonder: Why on earth did they do something like that? What motivates people who don’t need money to steal from others? Or, you read
in the newspaper that a man living around the corner from you has been arrested for growing marijuana in his garage You think back to the times you passed this man’s house and smiled a greeting at him And you wonder: Why didn’t I realize what he was up to? Who was he going to sell the mar-ijuana to anyhow? From these broad curiosities, you might begin to think about how these questions could be explored
or answered and how you might research these phenomena
Or you might think more generally about how we define particular forms of crime as “urban” as though they couldn’t occur in the suburbs, from which you might define research questions about why some people receive long prison sen-tences and others short ones for the same crimes
The preceding examples serve two important purposes First, they point out how ideas promote potential research endeavors Second, and perhaps more important, they suggest
in social scientific research similarly may seem obvious at
first, but they must always be clearly defined
Typically, concepts have two distinct parts: a symbolic
element (a word, symbol, term, etc.) and an associated
defini-tional element People learn definitions for certain concepts in
a variety of ways For example, children may learn the
con-cept of honesty explicitly when a parent or teacher specifically
instructs them on its meaning Or it may be learned
implic-itly through a more diffuse, nonverbal process of observed
instances in which either dishonest behavior is corrected or
honest behavior is rewarded (either through comments or
actions) In either case, eventually each of us comes to
appre-hend the meaning of honesty Yet, if asked to define it, people
may offer slightly different shades of understanding One
person might say, “Honesty is not lying to people.” Another
might offer, “Honesty is not taking property that belongs to
other people.” And a third individual might claim that
“hon-esty is being able to be trusted to do what you promise to do.”
Obviously, these responses suggest that even a fairly common
concept may have multiple meanings Each of these
defini-tions is valid on its own merits (some would say “true”) Yet,
they are different from one another and therefore each
defini-tion addresses only some small pordefini-tion of the larger concept
Unlike dictionary definitions, which are intended to cover all
known uses of a term, scientific definitions need to highlight
the (usually) single meaning that is pertinent to one’s study
In the social sciences, vague or unclear definitions create
enormous problems Specificity is critical when conducting
research Therefore, an important part of developing social
scientific theory is to first define relevant concepts that will be
used in a given research process or project
Indistinct, unclear, or vague definitions of concepts
create obstacles to the advancement of knowledge and
science After noting that there were many different
defini-tions in the literature for the concept gang, Richard Ball and
G David Curry (1995, p 239) explained the term carried
too many “latent connotations” to be treated as a single
thing By “latent connotations” the authors refer to the vast
world of conceptual associations that the term “gang”
car-ries While one researcher might describe a new pattern of
urban school kids grouping together for status and mutual
protection as “increasing gang presence in the schools,”
readers might well assume that gang presence means
weapons, drugs, fights, or the allegiance of school groups
to well-known regional gangs such as the Crips or the Latin
Kings Presumably, fewer people will assume that the term
refers to biker gangs or chain gangs But any vagueness
in the use of key concepts invites speculation The need
for this sort of specific definition of concepts will be made
clearer later in the discussion on operationalization
Concepts rarely occur in isolation Rather, they occur in
what Neuman (2000, p 43) refers to as concept clusters or what
we may call propositions One can connect different concepts
or conceptual thoughts to each other through propositions
Trang 26Designing Qualitative Research 25
to conduct new research just to show that it’s still the same Similarly, many sociology texts like this one have, for years, used presumably familiar examples of research questions pertaining to binge drinking on campuses or peer pressure
in high schools that we may have collectively contributed
to the impression that these are urgent social problems that require active research immediately Yet, unless you have something truly innovative to add to these frequently discussed subjects, there is little benefit to running around campus asking people how much they drink
So, you begin with an idea But how is this related to theory? Many research projects begin with formal state-ments of the ideas and theory on which the empirical
research is to be based This has been called the research model (Frankfort-Nachmias & Nachmias, 2007) This
theory-before-orientation has been nicely described by Karl Popper (1968), who suggested that one begins with ideas (conjectures) and then attempts to disprove or refute them through tests
of empirical research (refutation) And yet, theory is based
on data Research must occur before theory can be
devel-oped This research-before-theory orientation was expressed by
Robert Merton (1968), who emphasized that research was an integral part of every stage in the development and testing of theory In other words, research may suggest new problems for theory, require theoretical innovation, refine existing theories, or serve to challenge past theoretical assumptions.The approach offered in this book views theory-before-research and research-before-theory perspectives
as highly compatible, and most researchers move fortably between them Realistically, we often adopt an approach that encompasses both models The research process is conceived as spiraling rather than linear in its progression You begin with an idea, gather theoretical information, reconsider and refine your idea, begin to examine possible designs, reexamine theoretical assump-tions, and refine these theoretical assumptions and per-haps even your original or refined idea Thus, with every two steps forward, you take a step or two backward before proceeding any further What results is no longer a linear progression in a single, forward direction Rather, you are spiraling forward, never actually leaving any stage behind completely This spiraling approach is drawn in Figure 2.1
com-To simplify understanding of the individual elements
of this model as I discuss them, let’s redefine the stages slightly, as follows:
Ideas ➞ Literature Review ➞ Design ➞ Data Collection and Organization ➞ Analysis and Findings ➞
Dissemination
As illustrated, you begin with some sort of rough idea
for a research study The next stage in the process is to begin thinking and reading about the topical idea As you begin reading related and relevant literature on the topic,
a central research orientation that permeates this book This
orientation is the attitude that the world is a research
labora-tory and that you merely need to open your eyes and ears to
the sensory reality that surrounds all of us to find numerous
ideas for research In fact, once you become familiar with this
orientation, the biggest problem will be to filter out all the
many possible researchable ideas and actually investigate one!
Most experienced qualitative researchers will agree that
if you drop an investigator into any neighborhood, he or she
will manage to identify a research idea, develop a research
plan, and project potential research findings before lunch
I sit on a morning commuter train and look around me
The difference between the crowded rush-hour trains and
the sparsely populated later trains is extreme How did we
come to define “work hours” in such a regimented fashion?
How is this changing as more people are able to
“telecom-mute”? If the manufacturing sector is shrinking in the
United States, while service work is growing—and service
work is increasingly done around the clock—why is rush
hour still so crowded? And what about other parts of the
world where manufacturing is increasing? Are these places
experiencing greater rush-hour traffic than before? How
will they choose whether to build more roads for private
cars or more train lines for mass transit? And finally, why
do people making private phone calls in public places, like
trains, talk so much more loudly than everyone else? I could
spend the rest of my career trying to understand this train
This notion is likely to contrast dramatically with the
inexperienced researcher’s fear that he or she cannot even
think of anything worthwhile to research There may be
considerable truth to the optimistic view of experienced
researchers This does not mean, however, that all research
ideas will be equally easy or interesting to research
Some ideas will be more difficult to investigate than
others This is because those who control access to a
given location—what the literature calls gatekeepers—or
the subjects themselves may be reluctant to cooperate
Gatekeepers are discussed in greater detail in Chapter 6
Also, some ideas may initially seem extremely
interest-ing but become rather plain or uninspirinterest-ing on further
investigation Some ideas are interesting to think about
but impractical, unethical, or even impossible to study in
a rigorous fashion The impacts of emotional trauma, for
example, can be inferred through many case studies of
trauma victims, but you cannot test these inferences in an
isolated experimental setting without deliberately inflicting
trauma on your research subjects Some students
under-stand research in relation to findings that they have been
taught in other sociology classes For example, the research
question “Do advertisements represent women in a
sexu-ally exploitive fashion?” was once an important question to
look into Now, after years of study, we know the answer is
yes, and until something changes in the advertising field to
call that into question, it is much less useful or interesting
www.downloadslide.com
Trang 27are, the fastest way to immerse yourself in a new topic is still to spend a few hours pulling bound volumes off of shelves and browsing the most promising articles in them.The next task is to begin thinking creatively about cryptic subject topics related to your rough research idea
or question and to search for these topics in the indexes For the preceding example, you might make a list that includes “alcohol use,” “collegiate alcohol use,” “alcohol
on campus,” “drinking,” “males and alcohol,” ity,” “Americans and alcohol,” “social drinking,” “sub-stance abuse in college,” “campus problems,” and so forth
“masculin-It is important to develop a number of different subject areas to search Some will be more fruitful than others, and perhaps some will yield little information This is because both the print versions and computer-based versions of indexes are created by humans Because of this, indexes unavoidably suffer from the problem of terminological classification bias In other words, even though these indexes are cross-referenced, if you do not use the same term or phrase used by the original indexer, you may not locate the entries he or she has referenced Your search of the academic literature is guided by your research topic, but the literature search itself will help you to refine your questions Only after you have immersed yourself in what
is known about the topic, what is speculated about, and what is unknown can you define the useful angle for your study that can promise to make an actual contribution
A promising research project can be quickly derailed by
a weak literature review For instance, some years ago, Bruce Berg became interested in the idea of doing research about women in policing More directly, he was interested in the effect of policing on female officers He asked his graduate student to see if she could locate some material about female police officers (Getting your graduate students to do an ini-tial search is one of the most effective ways to begin a project.) When she returned the next day, she reported that there was virtually nothing in any of the index databases on the topic
“female police officers.” Berg asked if she had tried “women
in policing,” or “women police officers,” or even “minorities
in policing.” Sheepishly, she explained she had not thought
to do that and returned to the library When she returned,
you should also start turning this idea into a research
ques-tion or even a set of researchable foci As suggested by the
fluidity of the spiraling approach offered in this chapter,
your research idea should flow into a potential research
question that may continue to shift, change, and take
form as the research process unfolds Even though your
research question(s) may change as you proceed through
the research process, it is important to establish a focus
for your research question or a series of research aims
2.3: Reviewing the Literature
2.3 Describe the importance of authentic literature in
research
After developing a rough idea for the study, you will need
to begin examining how others have already thought
about and researched the topic Let’s say an idea for some
research begins with an interest in alcohol use by male
col-lege students, despite my warnings that this ground has
been covered extensively already You might formulate a
rough question for research such as the following: What
is the relationship between college and drinking among
American males? This rough idea already shows elements
of refinement It has been limited to consideration of only
American males But it is still very general and unfocused
The next step is to visit the library or its Web site to get
started on a literature review Because every library is
different, you will need to familiarize yourself with the
sorts of databases, periodicals, and books that are readily
available to you Most periodicals are available to browse
online through databases such as Infotrac or Research
Navigator’s ContentSelect, but for books you have to
actu-ally go to a building Some libraries have subscriptions to
many journals, but not all of these may be useful for social
science research, let alone a specific topic such as alcohol
drinking by American male college students Different
libraries also provide different methods for accessing
materials, including large selections of in-print periodicals
maintained both in current stacks and in bound versions
in back stacks or in the open library As convenient as pdfs
Ideas Theory Design CollectionData Analysis Dissemination
Figure 2.1 The Spiraling Research Approach
Trang 28Designing Qualitative Research 27
segments of the information are reproduced, and one must still acquire the actual text from the library or through a purchase And unlike scientific research tools, Internet search engines retrieve far more information that is of pos-sible general interest but mostly useless in formal research For example, access the Internet and try running a search for the term “concept.” The initial results may be less than useful if you are writing a scholarly term paper, article, research report, or proposal
We need to make an important distinction here between the Internet as a document delivery service and the Internet as a document repository In the first case, the traditional materials of basic research—peer-reviewed sci-entific articles—may be downloaded via the Internet right
to your computer The source of the materials is the journal
in which it was first published, whether you got your copy
by photocopying, downloading, or from a published reader (e.g., Lune, Pumar, & Koppel, 2009) The Internet just gets you the article faster In the second case, however, the mate-rials were actually published on the Web and can only be accessed through an Internet search As a very general rule
of thumb, the first set of materials is valid and useful while the second is suspect and unreliable Reviewing the litera-ture in a field of study means reading valid research, not abstracts, blogs, magazine articles, rants, or encyclopedias
We take the Internet for granted, and such placency with this technology can be dangerous for a researcher Yes, the Internet is enormously fast, and yes,
com-it has evolved in less than three decades to provide access
to many millions of documents However, the quality and integrity of all the available documents are not equal The
Internet epitomizes the concept of caveat lector—Let the reader beware.
The Internet allows you to access information from
a variety of governmental and private sources, as well
as from online electronic journals, books, ies, archives, and even newspapers Most governmental agencies have Web sites that offer the public copies of recent (and often backlogged) reports, pamphlets, news releases, and other forms of information There are also Web sites, however, that offer inaccurate, erroneous, or fabricated information I once had the unpleasant experi-ence of reading a student “research” paper on homosexu-ality in America that was entirely based on information
commentar-he had downloaded from a couple of hate-group sites Amazingly, the student had (apparently) skimmed the materials so carelessly that he accepted their claims as established facts without even noticing the death threats, support for Nazi extermination programs, or frequent use
of curses and other invectives He hadn’t realized that the sites were not valid and reliable sources of data Granted, this is an extreme example: sort of the Internet-age ver-sion of writing your term paper on the bus ride to school
on the morning that it’s due With just a little care, this
she was carrying a list of literally dozens of references I have
seen many instances of similar thinking among students
who are first learning to conduct research Returning to the
preceding example, many of my past students have proposed
research on male college drinking only to declare that there
is virtually no literature on “campus drinking by men” or
“why men in college drink.” Yet, using the separate searches
mentioned earlier would yield thousands of relevant articles
The lesson to be learned from this is that you must not be too
restrictive in your topics when searching for reference
materi-als in indexes In fact, most online indexes provide users with
a thesaurus to assist them in locating subject terms used to
index material in the database
When beginning your literature review, it is no
lon-ger necessary to arrive at your library empty handed and
hoping to stumble across good materials Library
cata-logs, database search engines, book reviews, and journal
tables of contents are all available online and may be
scoured for promising sources from the comfort of your
own coffee shop The majority of academic articles may
be downloaded in pdf format depending on your library
subscription services You can pore through these more
immediately accessible works, saving your actual visit
for older or harder-to-find books and articles Still, there
is much to be gained by casual browsing in the library
stacks Search engines, databases, and the vast information
available via the Internet are wonderful tools and places to
begin searching for literature They can provide enormous
amounts of information But they only give you access
to the information that someone else has already added
to the pertinent databases Frequently, however, there is
no substitute for physically thumbing through journal
indexes It is also important when using the Internet to be
careful about the legitimacy of materials taken from the
Web, which we will now consider in detail
2.3.1: Evaluating Web Sites
In the years since the first edition of Qualitative Research
Methods for the Social Sciences was published, Internet
searches have become the first, and often the only,
informa-tion source for many millions of users, including
profes-sional researchers Google even provides separate search
levels called Scholars and Books We strongly endorse,
and rely on, these different tools, but they are not the
sole source of literary materials a good researcher should
employ Google Scholar, for example, is full of papers and
articles that can be downloaded in their entirety;
unfortu-nately, many of these require a fee or membership in some
sort of literary subscription Google Books allows one
to explore thousands of books—but not in their entirety
Sometimes, the topic one is seeking does yield enough
information to be used, and the full citation information
is provided in the search However, at other times, only
www.downloadslide.com
Trang 29you get many hits Do not use only the first one you find Carefully check a number of comparable sites to ensure the information is comparable If you find that there are glaring contradictions or discrepancies, you should be very cautious about using this information.
2.3.2: Content versus Use
By now, you should have begun to amass a large quantity
of documents to include in your review of the literature Naturally, you will need to begin taking some form of notes on the various pieces of literature you have obtained There are a number of ways you can keep such records and notes What follows are a few general suggestions for organizing your work There are no rules, however, and you will do best to discover the style that works best for your own ways of thinking
It is difficult to educate yourself on a new area of study while also learning who the key authors are in this area while also becoming familiar with the specialized vocabulary of research on the topic while thinking about the meaning of the findings presented while planning the paper that you will write It helps if you can break the work down into different parts I prefer to maintain a strict distinction between two questions: What does the mate-rial say? And how does this relate to me? In other words, taking notes on the content of the literature you study is distinct from taking notes on how to use that literature in your own work
Writing notes on the content of research articles and books is a lot like preparing a junior high school book report First, record the full citation information for the article or other source Next, identify the major claim(s), methods, and subject matter of the work Under that, begin to write out all of the best parts—the quotable expla-nations, definitions, and findings that make this work unique Quote each exactly, with quotation marks, and note the page numbers When you are done, you should have a brief file that encapsulates the key parts of your source, making it much easier to draw on when you write Chapter 12 discusses the problems with paraphrasing and with careless use of quotes in the section about plagiarism There are other benefits to careful quoting
Copying over exact quotes often seems tiresome and unnecessary Since we are primarily interested in ideas, not phrases, one might think that a paraphrase is better I rec-ommend otherwise If you, as an investigator, paraphrase material in your content notes, it is possible that you might slant or alter meanings Without intending to, you might have misread, misinterpreted, or poorly paraphrased material When you go through the notes looking for agreement among authors, you might find paraphrased statements that seem to represent similar ideas, but that actually do not accurately represent the sent iments of the
error would never have occurred But other errors may be
harder to detect It is critical that you carefully evaluate
documents before quoting them Here are a few questions
you might want to consider before accepting information
from a Web site as valid:
1 Whose Web site is it? Before you even start to
con-sider the veracity of the text on a particular Web site, look
at the URL to get a sense of the authenticity of the material
on that site Personal pages are not necessarily inaccurate,
but you should nonetheless consider the authority and
expertise of the author very carefully Just about anyone
with a computer can launch and maintain his or her own
Web site When you consider using information taken
from an individual’s personal Web site, you still should be
cautious and consider the credibility of the individual or
group that is operating and maintaining the site
2 What is the nature of the domain? The domain
repre-sents a kind of hierarchical scheme for indicating the
logi-cal and sometimes geographilogi-cal venue of a Web page In
the United States, common domains are edu (education), gov
(government agency), net (network related), com
(com-mercial), and org (nonprofit and research organizations)
Outside the United States, domains indicate country: ca
(Canada), cn (China), uk (United Kingdom), au (Australia),
jp (Japan), fr (France), and so forth Is this an official
gov-ernment Web site or that of a well-known and reputable
organization? Is it operated and maintained by a private
group that has a special purpose or motive for having the
site and offering the materials you are considering? As I
mentioned earlier, there are a number of Web sites
spon-sored by hate groups The information offered on such
sites may sound like the reports of scientific studies, and
the reports and documents may even look official Yet,
much of the information on these sites is likely biased
and designed to be self-effacing and positive in order
to sway readers to think favorably about the group’s
viewpoints
3 Is the material current or dated? You should check to
see how frequently the Web site is updated If the
mate-rials have not been updated recently, you may want to
question how reliable a source it is Consider also whether
links are active or have expired or moved Naturally,
just because a site is well maintained and information is
regularly updated doesn’t mean it is necessarily a good
site in itself, and some material may not require constant
updates However, issues of currency are important when
conducting research and should be considered when
eval-uating information taken from a Web site
4 Can the information be corroborated? Sometimes the
material you find on a Web site seems odd or unusual,
and further investigation suggests that it may not be
truthful When this happens, do not use it Often when
you undertake a search using an Internet search engine,
Trang 30Designing Qualitative Research 29
material for publication, but I have no usable sources for any of my claims
Fortunately, there are technological solutions for those of us too rushed or too lazy to write everything down Most of the databases that you might use to find many of your materials—whether books or articles—will also allow you to save the complete citations in any of the standard writing styles And many will generate records suitable for a bibliography program Bibliography soft-ware is extremely useful for storing accurate and com-plete lists of materials you have read, whether you ended
up using them in your current paper or not They also allow you to store keywords with each record, which we know is helpful And since you can download the cita-tions and copy them into files with a few keystrokes, you have little opportunity to introduce typos Your univer-sity library may offer free or reduced-cost software for this, and many programs can be downloaded for little or
no money anyway You can try out a few and decide for yourself
First, though, we need to think about how we use all
of these notes
New work is built on a foundation of old work We take the best of what is currently known and weave it together to form the solid ground on which to place our own, new, contributions The content notes that I described earlier are not such a foundation To push the metaphor a little more, they are the materials from which we construct that foundation
Let’s imagine that I am starting a study of teen drug use Clearly, some of my background literature would come from the field of juvenile delinquency studies, from which
I would learn of the statistical distributions of different forms of youthful criminal behavior, the nature of interven-tions and their success and failure rates, and criminological theories for such behavior All of this is a start, but little of
it would be exactly on my topic The youths I’m studying aren’t necessarily thieves or thugs, gang members, or even dropouts Most of them are probably suburban stoners But the delinquency literature is one pillar
There is a rich social-psychological research literature
on adolescence One can get lost in such a broad field, soaking up thousands of pages of new information For the sake of efficiency, I would need to limit my reading with the strategic use of additional keywords I would obviously read about teen drug use, and teen drinking and probably teen smoking as well This body of research would provide another pillar, with theories and data about the nature and causes of adolescent behaviors that are viewed as “antisocial.” Notice that “antisocial” behavior will overlap with some of what the delinquency literature calls “criminal” behavior Relating the two to each other,
or separating them in a useful way, is part of my job as the writer of my own research paper
original authors Using verbatim excerpts ensures that this
will not occur Either the authors did say similar things
or they did not Also, block copying from pdfs into a
word processor is faster and more accurate than typing it
yourself
I also recommend saving keywords with each file to
describe the content It may seem like extra work at the
time, but it can be invaluable later when you need to find
all of your sources on antidrug laws, or to locate that one
piece you vaguely remember containing the story about
the homeless dog If it’s possible, it also sometimes helps
to make liberal use of subfolders to store your notes
Under the “social movements” folder, I might have folders
for “American” and “European” cases, or “cultural”
move-ments in one and “material” goals in another Of course
the problem there is that you could have a European
cul-tural movement that is pursuing the expansion of access to
things of material value, in which case you could file that
almost anywhere This is why keywords are often more
useful ways to identify source files
With keywords, you can very quickly sort the
sum-maries into different categories as you need them (e.g.,
placing all the notes about police detectives together, or
all the theory pieces in one place) In this manner, you can
assemble the material into an organized sequence that
will reflect how you plan to write the report or paper This
allows you to read through the relevant materials for each
section rather than repeatedly read through all of the
mate-rial in order to write a single section
Keyword searches also allow you to assess whether
multiple authors actually have made similar statements
about issues or situations In turn, you are able to make
strong synthesized statements regarding the work or
argu-ments of others For example, you might write, “According
to Babbie (2007), Frankfort-Nachmias and Nachmias
(2007), and Leedy and Ormrod (2004), the design stage
is a critically important element in the development of
a research project.” Making such a synthesized
state-ment, which collapses the arguments of three individuals
into one, can be easily accomplished because you would
have notes for each author conveying this sort of general
sentiment
I have violated all of this advice at times, and so I have
learned the hard way about the importance of good record
keeping Before we all had laptops, I had actual folders
with pieces of paper in them to store my notes To save
time, I would write the author’s name on the top of a note
sheet without writing down the title Weeks later, after I
had inserted a great quote from “Smith” into my paper,
I would have to take it out again because I was unable
to figure out if this was Dorothy Smith (1987), Michael
Peter Smith (1998), or someone else altogether I still
have a folder containing an entire conference presentation
without a single citation in it I would love to rewrite the
www.downloadslide.com
Trang 312 Explain competing conceptual frameworks Some drug
use studies center on the issue of blame Are the users bad people? Are their parents so? Have their schools failed them? Other studies look at control efforts, police budgets, the availability of treatment options, and enforcement policies So, one set of readings is concerned with the problems of supply, while others are all about demand
3 Clarify the focus of your own work I might, for example,
explain the unique features of a symbolic ist approach to state that I am interested in under-standing the meaning of the act (drug use) from the perspective of the user, and not from the perspective
interaction-of parents or politicians
4 Justify assumptions Drug use patterns are cyclical The
popularity of specific drugs rises and falls endlessly
By using government data on drug sales and arrests, I can back up my claim that declines in use of one drug are usually accompanied by increases in the use of others Therefore, I might reject a local mayor’s claim that his own policies toward drug control are respon-sible for the recent decline in whatever drug is going out of favor
The main point is that your literature review section
is like an essay on the background to your topic It has an introduction, in which you explain what your topic is and what you are reviewing It has a point, which is to support your research question and your design There is the body
of the paper, in which you present the information that defines the background to your work Therefore, you can start with an outline as you might for a larger paper And this is where you start to map out a strategy for putting your content notes to use You can lay out the major claims
of the literature, decide what order to address them in, and begin to write out notes about what you want your readers
to understand about the material Ultimately, you would produce a coherent essay that flows from the introduction
to the conclusion, touching on the various works of the field along the way
Returning to the example above, my written ture review on drug use might emphasize the transitory nature of most use, in contrast to the literature on addic-tion I would emphasize the situationally specific nature
litera-of much use and include references to research on how and when people stopped using whatever they had been using These references to research findings would include citations to the sources of the information But the writing
is about the findings, not the sources Few things are as boring as a list of things other people have said You may have an early draft of your paper that says, “researcher
A looked at smoking practices , but researcher B found otherwise In researcher C’s study, ” But don’t hand that in The final version should contain a paragraph or
A third pillar for this work might come from research
on families There might be household-level data that I
would want to consider Of course, the drug of choice
among youths varies by socioeconomic status Powdered
cocaine is more popular among people who can afford it,
while crack cocaine is accessible to low-income
consum-ers Heroin goes in and out of fashion, while marijuana
remains the perennial favorite among casual users I
would certainly want to know more about who is
typi-cally using what in order to both plan and describe my
research
Finally, at least for purposes of this discussion, there are
classic works that simply have to be included if I’m going
to make any sort of conceptual argument about my topic
If I want to investigate youth drug use in relation to anomie,
then I will have some discussion of Durkheim If I want to
address the social context in which the drugs are used, or
the meaning of the act to the users, then I would certainly
start with Norman Zinberg’s (1984) Drug, Set, and Setting.
With all of this research literature consumed and
reduced to notes, I have my materials But I still don’t
have my foundation Simply listing all of the different
viewpoints that all of this past work has claimed or
dem-onstrated would produce more confusion than clarity
Results in one source, taken at face value, contradict the
results of another Each of the sources addresses some
small part of my study, but none of them directly answer
my question (Notice that if one of them did answer my
question, and I accepted that answer as valid and
com-plete, then there would be no justification for me to do my
work at all We’re supposed to use our work to go beyond
our sources.) So how do I use my notes?
Let’s recall the purpose of writing a literature review
You provide the background needed to educate your
read-ers enough so that they can undread-erstand and follow what
you are doing and so that they can appreciate the need
for your work The review of past research brings them
up to speed, introduces and explains the major concepts
with which you are working, does not introduce concepts
that you don’t need, and provides the motivation for your
new research (Galvin, 1999) Ideally, by the time
individu-als have finished reading your background section, they
should be on the edge of their seats wanting to know what
you have found
There are many ways to write a literature review
sec-tion A few of the things you might try to do when writing
yours are as follows:
1 Dispel myths One of the myths of drug use is that we
could eliminate it entirely if we had just the right
poli-cies and strategies Yet, studies indicate that drug use
is universal, across all sorts of times and places, under
all regime types, and through all kinds of economic
and social conditions
Trang 32Designing Qualitative Research 31
Let us return to the earlier research idea: What is the relationship between college and drinking among American males? After reading through some of the lit-erature, you might begin to refine and frame this idea as a problem statement with researchable questions:
Problem statement This research proposes to examine
alcohol-drinking behaviors in social settings among college-age American males
Research questions A number of questions are addressed
in this research including (although not limited to) the following:
1 What are some normative drinking behaviors of young adult American males during social gather-ings where alcohol is present?
2 How do some young adult American males age to abstain from drinking (e.g., avoidance rituals) while in social situations where alcohol is present?
man-3 How do young adult American males define priate drinking practices?
appro-4 How do young adult American males define lem drinking?
prob-These questions did not just happen spontaneously They were influenced by the literature about drinking practices among Americans They resulted after the inves-tigator began thinking about what issues were important and how those issues might be measured This required the researcher to consider various concepts and definitions and perhaps to develop operationalized definitions
2.5: Operationalization and Conceptualization
2.5 Describe the process of operationally defining a concept
When someone says, “That kid’s a delinquent,” most of
us quickly draw some mental picture of what that is, and
we are able to understand the meaning of the term quent.” If, however, someone were to ask, “How would you define a delinquent?” we would probably find that some people think about this term differently from others For some, it may involve a youth under the legal age of adult jurisdiction (usually between 16 and 18 years of age) who commits law violations (Bynum & Thompson, 1992) For others, a delinquent may be simply defined as a youth-ful law violator (Thornton & Voigt, 1992) Still others may require in their definition some notion of a youth who not only breaks a law but also is convicted in court of this law violation (Siegel & Welsh, 2008) In other words, there are
“delin-a number of possible definitions for the concept delinquent.
more on smoking practices as they apply to your topic
(with a parenthetical, in-text citation to researcher A)
Further research may raise questions about how applicable
that is (B)
Notice how completely unlike a junior high school book
report this final essay is No one, honestly, no one wants to
read your content summaries Your papers are not
strength-ened by a long diversion into listing a bunch of things that
you have read All of that content summary was for you, to
make it easier for you to write the real literature review part
TRYING IT OUT
There are a number of ways you can practice aspects related to
the planning of research The suggestions below should provide an
opportunity to gain some experience Although these are useful
experiential activities, they should not be confused with actually
conducting research.
Suggestion 1
Look for two reports on marriage and family life in Asia using any
search engine Evaluate these reports using the criteria described
on page 28:
a Whose Web site is it?
b What is the nature of the domain?
c Is the material current or dated?
d Can the information be corroborated?
Reflect on your findings from both these reports and explain which
one you think is more authentic
Suggestion 2
Design a literature search on the topic “trends in high school
dropout rates around the world.” Identify the general areas that
your background literature should relate to and the topics it should
incorporate Then make a list of keywords you would use to
search for relevant literature under each of these topics.
2.4: Framing Research
Problems
2.4 Give an example of a problem statement with
researchable questions
Research problems direct or drive the research
enter-prise How you will eventually conduct a research study
depends largely on what your research questions are It is
important, therefore, to frame or formulate a clear research
problem statement Remember, the research process begins
with an idea and only a rough notion of what is to be
researched As you read through and collect information
from the literature, these rough questions become clearer
and theoretically more refined
www.downloadslide.com
Trang 33course, this technical definition tells us nothing about the socially relevant concepts of “ideal weight,” or “under-weight,” or “overweight,” and is clearly insufficient to talk about body image and the marketing of weight-loss pro-grams There are medical definitions of “obesity,” which inform our social definitions of the concept, up to a point
As social researchers, we also need to know where the perceptions of obesity and the impact of obesity on other behaviors depart from the medical measures of one’s body-mass index However, on its own, removed from any social context, we know what weight is
Unfortunately, not all concepts are as easy to define
as weight or as easy to measure Polit and Hungler (1993) and Polit and Beck (2007), for example, suggest that many concepts relevant to research in nursing are not operation-alized simply For instance, in nursing research, the quality
of life for chronically ill patients may be defined in terms
of physiological, social, and psychological attributes If the nurse researcher emphasizes the physiological aspects of quality of life for chronically ill patients in his or her defi-nition, the operationalized component may involve mea-suring white blood cell counts or oxygen output, assessing invasive surgical procedures or ventilation procedures, measuring blood pressure, and so forth
If, on the other hand, quality of life for chronically ill patients is defined socially, the operationalized elements
of the definition would need to measure family or social support, living arrangements, self-management skills, independence, and similar social attributes, as well as experiential measures such as pain, mobility, depression, and sense of self Likewise, if the nurse researcher uses a more psychological conceptualization, the operationalized measures would be directed along the lines of the patients’ emotional acceptance of chronic illness
Let’s try another illustration of defining and tionalizing Say you are interested in studying to what degree or extent people are religious To begin, you must
opera-define the concept religious For this example, religious
will be defined as how actively one is involved with his
or her religion In a sense, we would wish to know how important religion is to one’s life on a daily or larger basis Next, you must decide what kinds of information inform others about someone’s active involvement in religion After consulting the literature, you decide that you know how religious someone is by knowing whether that person believes in a divine being, attends organized religious services on some regular basis, prays at home, reads religious materials, celebrates certain religious holidays, readily declares membership in a particular religion, par-ticipates in religious social organizations, and contributes
to religious charities
In effect, you, the researcher, are saying, “I can’t immediately apprehend a person’s religiousness, but I can think about what elements seem to go into making
If you, as a researcher, are interested in studying the
behavior of delinquent girls, you will first need to clearly
define “delinquent.” To ensure that everyone is working
with the same definition and mental image, you will need
to conceptualize the term and operationalize its measurement
This process is called operationally defining a concept.
Conceptual definitions of a term limit our approach to
the particular perspective that we are taking for purposes
of our study A well-defined set of conceptual definitions
reveals and justifies the specific kinds of data we will need
to answer our research questions Operational definitions
concretize the intended meaning of a concept in relation
to a particular study and provide some criteria for
mea-suring the empirical existence of that concept
(Frankfort-Nachmias & (Frankfort-Nachmias, 2007; Leedy & Ormrod, 2004) The
operationalization tells us how we will get that data
In operatively defining a term or concept, you, as a
researcher, begin by declaring the term to mean whatever
you want it to mean throughout the research Although
it is important for your readers to understand what you
mean when, for example, you use the concept delinquent,
they need not necessarily agree with that definition As
long as they understand what you mean by certain
con-cepts, they can understand and appraise how effectively
the concept works in your study You may even use your
literature review to introduce other working definitions
of the concept in order to distinguish your definition from
these Researchers routinely introduce changes in the
for-mal definitions of their concepts, building on the
accom-plishments of past work, refining our tools in pursuit of
the nuances of our study topics Different definitions may
coexist, each highlighting a particular aspect of the
con-cept The important thing is to let your readers know what
you mean when you refer to the concept
Once defined, the concept needs some way to be
measured during the research process In quantitative
research, this means creating some index, scale, or
simi-lar measurement indicator intended to calculate how
much of or to what degree the concept exists Qualitative
investigators also need agreement over what a concept
means in a given study and how that concept is to be
identified and examined How will the researcher gather
empirical information of data that will inform him or her
about that concept?
Consider, for example, the concept weight As a
researcher, you might define the concept weight as the
amount of mass an object possesses in terms of pounds
and ounces (measured at g0) Now everyone holds the
same concrete meaning and mental image for the concept
weight How shall we measure weight as a social concept?
Operationally, weight can be determined by placing an
object on a scale and rounding to the nearest ounce This
operational definition clearly tells others what the concept
is designated to mean and how it will be measured Of
Trang 34Designing Qualitative Research 33
interviews.” Really, it only states that sampling matters It doesn’t tell us how to define the community, what makes people representative, or even how many subjects to aim for In contrast, the statement, “I will identify two key informants from the neighborhood, and, using snowball sampling, recruit 15 subjects from each informant’s per-sonal network” defines goals and criteria for inclusion It tells the reader what you actually intend to do
The most basic elements of research design involve conceptualizing the things you need to know, operation-alizing the data collection that will tell you those things, and applying this data to your testable propositions Each stage of research design involves a series of impor-tant decisions about the research idea or question(s) What types of information will be gathered and how will it be measured? This part of the design specifies the kind of data you need to answer your question or test your thesis Where will the research be undertaken, and among what group or groups of people (questions of site, setting, and sample)? This part specifies how you will get the data that you just described as necessary Finally, the design must include a data-analysis plan Once you have the data you want, how will you use it to answer your question?
There are many other considerations to take into account along the way You must decide whether to use
a single data-collection strategy or to combine several strategies (data triangulation) Will you undertake the study alone or with the assistance of others (multiple investigator triangulation)? You must consider whether the study will be framed by a single overarching theory or
by several related theories (theoretical triangulation) How much will the project cost in terms of time and money, and how much can you actually afford? Are the data-collection strategies appropriate for the research questions being addressed? What will the data (physically) look like once they have been collected? How will the data be organized and analyzed?
In effect, during the design stage, you, the tor, sketch out the entire research project in an effort to foresee any possible glitches that might arise If you locate
investiga-a problem now, while the project is still on the drinvestiga-afting board, there is no harm done After the project has begun,
if you find that concepts have been poorly conceived, that the wrong research questions have been asked, or that the data collected are inappropriate or from the wrong group
of people, the project may be ruined
In addition, the researcher must consider what Morse
and Richards call the pacing of the project By pacing, Morse
and Richards (2002, p 66) mean planning the sequence
of various components of the study and the movement between data gathering and data analysis This planning requires considerable decision making during the design stage and the flexibility to make additional changes during
up or representing observable behaviors that I
under-stand and associate with the meaning of religious.” By
obtaining information regarding the subset of observable
attributes delineated earlier to represent religious, you
can study the concept of religiousness, or religiosity As
you think about the observable attributes of religiosity—
or of any other concept—you should again peruse the
literature By spiraling back into the literature stage, you
can seek ways in which others have examined the
con-cept You may borrow some of these previous attributes,
or you may create others
In some forms of qualitative research, the investigator
is not as rigorously concerned with defining concepts in
operational terms as outlined here This is because some
forms of interpretative and phenomenological research
seek to discover naturally arising meanings among
mem-bers of study populations However, in many cases of
qualitative research, failure to define and operationalize
concepts will spell disaster If, as a researcher, you have not
made clear what your concepts mean, your results may be
meaningless in terms of explanatory power or
applicabil-ity If you have not thought about how data will be
col-lected to represent attributes of the concept, it will be very
difficult for you to determine answers to research
ques-tions And if you have not worked with the literature in
developing relevant meanings and measurable attributes,
it will be impossible for you to see how eventual results fit
into this extant body of knowledge
Your next problem, then, is to determine exactly how
information about various attributes will be obtained As
you reach this point, you move one foot forward toward
the design stage of the research enterprise Naturally, your
other foot will remain in the literature stage
2.6: Designing Projects
2.6 Examine how the technique of concept mapping
assists the research design process
The design for a research project is literally the plan for
how the study will be conducted It is a matter of thinking
about, imagining, and visualizing how the research study
will be undertaken (Green & Thorogood, 2007; Leedy &
Ormrod, 2004; McTavish & Loether, 2002) Or as Valerie
Janesick (1994, 1998) metaphorically describes: design is the
choreography that establishes the research dance.
It’s often unclear what information goes into a
research design, and at what level of detail One possibly
useful rule of thumb is that you should write your plan
with enough specificity that you could turn it over to
several different people to implement, and they would all
be able to do more or less the same study For example,
consider a research design that stated, “I will recruit a
rep-resentative sample of subjects from the community for the
www.downloadslide.com
Trang 35represent propositions or elements of meaning Figure 2.2
shows a simple concept map for considering a theoretical framework for a study on health professionals’ perceptions
of obese patients
As Figure 2.2 illustrates, a concept map provides a means for organizing and thinking about the researcher’s notions about some subject or theoretical premise in a graphic or pictorial manner This tool is particularly use-ful for social scientists in developing and detailing ideas and plans for research It is especially valuable when researchers want to involve relevant stakeholder groups
in the act of creating the research project, as when
con-ducting participatory research efforts (see Action Research
in Chapter 7) It should be noted that typically one does not draft and complete a concept map all in one setting Even the draft of the concept model shown in Figure 2.2 is largely a first draft that could be refined as the researcher-developed additional information or narrowed his or her focus on specific issues How then, you may be asking, do you go about creating a concept map?
2.6.2: Creating a Concept Map
To create a concept map, you should first read widely on your subject; in short, begin examining the literature and amassing relevant documents on the topic As you read through these documents, you should also begin to keep
a record of about 10 or 12 key concepts or ideas Once you have identified these concepts, you may follow these sev-eral steps to create a concept map:
Step 1: List out the concepts on one page I use my laptop
for this, but some people are more tactile and fer to use post-it notes or small pad pages, writ-ing a separate concept on each pad sheet or post-it page The medium isn’t important, but it is impor-tant to be able to look at and move all of the con-cepts at once This step should yield a good-sized bunch of individual concepts
pre-Step 2: Rearrange the concepts on the page so you move
from the most abstract ideas to the most specific ones
Step 3: Now, move the concepts on the page under
sepa-rate columns, or create sepasepa-rate piles of notes so that ideas go directly below other related ideas This stage gives you a physical layout that represents your con-ceptual arrangement of the parts At this juncture, you also want to add additional concepts or labels that help to explain, connect, or expand the columns
or piles of ideas you are creating
Step 4: At this point, you can move the columns into
clusters of ideas located at some distance from each other, such that you can draw lines from the larger or broader concepts to the more specific and focused concepts and ideas This allows you
the course of the research: Once you select a data-collection
strategy, say field observations, when do you start? Once
you have begun, when do you stop? Should you include
interviews along with your field observations, even though
you did not originally plan to do so? All of these decisions
affect the pace, duration, and design of your research
Researchers in the social sciences typically conduct
research on human subjects The design stage is a time
when you, the researcher, must consider whether
ethi-cal standards and safeguards for subjects’ protection are
adequate You must make certain that subjects will be
protected from any harm Chapter 3 discusses issues of
research ethics in detail For now, regard the design stage
as the time when ethical proprieties such as honesty;
open-ness of intent; respect for subjects; issues of privacy,
ano-nymity, and confidentiality; the intent of the research; and
the willingness of subjects to participate voluntarily in the
study are appraised
2.6.1: Concept Mapping
For many researchers, the development of a research
design, creation of a theoretical framework, or even
devel-opment or use of existing theories can be a very
daunt-ing task At this juncture, therefore, I want to introduce a
tool that can assist you in this process and that can also
clarify confusions about a particular research design plan
or theoretical framework you may want to use This tool
is referred to as concept mapping or occasionally mind
mapping (Kane & Trochim, 2006; Maxwell, 2005; Novak,
1990) A concept map is a technique that allows you to
bet-ter understand the relationships between ideas, concepts,
plans of action, and the like by creating a pictorial
repre-sentation of these ideas, or plans, and their connections
Concept maps allow you to visualize specific connections
between ideas or activities you are thinking about, or to
connect new ideas to knowledge that you already
pos-sess about a theory or concept In effect, a concept map
permits you to better organize your ideas and plans as
you develop your research design or theoretical frame It
is quite literally your drawing board for working through
research and theoretical plans
Most sources suggest that the original idea of
con-cept maps can be traced to the work of Joseph Novak
(Novak, 1990; Novak & Gowin, 1995) and his colleagues
at Cornell University during the 1970s—first to explore the
way students learned science and then as a tool for
teach-ing science (Maxwell, 2005; Walker & Kteach-ing, 2002) To the
casual observer, a concept map looks like a pretty standard
flowchart; it is drawn with boxes or circles called nodes,
and connections between various nodes are represented
by lines, and sometimes arrows, and labels that identify
what each node is and what the relationships are as
repre-sented by the lines Together, these nodes, lines, and labels
Trang 36Designing Qualitative Research 35
the relationships you see between these clusters of concepts or ideas
Step 6: You may want to separately describe examples,
or even illustrations (pictures, cartoons) of actions that belong with and may illuminate the concepts and concept clusters
Step 7: Now, you should reorganize the concepts so that
the relationships among them are visually ent You may want to create a flowchart using
appar-to view where your tight clusters of ideas separate
from the looser, more distant interrelations
Step 5: You are now ready to begin the process of
mak-ing sense of the clustered ideas and connections
you have created in the previous steps In doing
this, you should again review your literature and
then begin to assign descriptive labels to the
con-nections among the clusters of concepts or ideas
These terms and labels should explain or identify
health Professional Perceptions of obesity
Social Perceptions of obesity
obese teen is observed
• Obese and normal weight people should be concerned about their weight
• Concerned with media’s effect on obese patients’ sense of self
• Obesity has serious ramifications for heart disease, diabetes, respiratory illnesses and asthma, and musculoskeletal disorders
• Recommend weight reduction of obese teens
• Recommend regular exercise regiments
• Seek to make teens heart smart
• Discuss self-image
• Teenage teasing of obese kids—especially girl
on girl
• Obese child avoids public/social settings
• Parental sarcastic comments
• Parental refusal to purchase clothes
• Parents leave diet materials around
• Possible changes in social perceptions via media
• Educational efforts in schools (both health issues and teasing/bullying discussions)
• The media has a strong negative
effect on obese teen’s sense of
self
Figure 2.2 Concept Map of Concepts Pertaining to Social and Health Professional Perceptions of Obese Teens
www.downloadslide.com
Trang 37in the state of whatever B is There are, therefore, three things to operationalize and measure: the variable A, the variable B, and whatever variables represent the rela-tionship between them.
How we proceed depends on the nature of our research question If, for example, we only want a general and descriptive measure of whether A influences B, we might design a large-scale possibly quantitative study with a lot of data for both A and B, with which we can tell whether they correlate or not But if the fact of this rela-tionship has already been observed, then we will probably want to know how and why In this case, neither A nor B are the key elements The value of our study will depend
on what we conceptualize for the arrow relating the parts and how we operationalize that
An example might help We have reason to believe that wealthy people vote more regularly than the work-ing poor (or presumably the nonworking poor) So X is wealth, and Y is voting But what is the “➞” about? One possibility is that people with less money feel more alien-ated from the political system overall and do not believe that their votes will make a difference In this model, the casual link is attitudinal, and we would have to opera-tionalize measures for confidence, connection, alienation, expectations, and the like Another possibility is that some portions of the population are encouraged to vote, such as car owners who receive registration forms through their motor vehicle departments, while others face barriers, such as densely populated voting districts with relatively few voting machines In this model, the causal link is structural, and our investigations would need to measure experiences and perceptions of those experiences The pic-torial representation of the relationships among the parts translates directly into our necessary conceptualizations, and hence, our data-collection plan
2.6.4: Setting and Population Appropriateness
During the research phase of a project, the tor must consider a rationale for identifying and using
investiga-a pinvestiga-articulinvestiga-ar setting investiga-as investiga-a dinvestiga-atinvestiga-a-collection site (Minvestiga-arshinvestiga-all
& Rossman, 2006) Data collection does not always involve recruiting random subjects and generally requires researchers to go out into the field and become involved with a population or group Decisions must also be made regarding who will collect the data and who or what will comprise the research study population The study might require data from a set of individuals Or you might collect data on organizations, communities, buildings, events, or nations While choices may be numerous, some advice is
in order First, it is best to be practical Select a site or ting that is reasonable in size and complexity so the study can be completed within the time and budget you have
set-various shapes (circles, squares, rectangles, etc.) to
depict the arrangement of the concept and/or idea
clusters and connective lines, as in Figure 2.2
Step 8: The final step is really a refining stage You may
want to show your cognitive plan to others
knowl-edgeable about the general subject area or others
working on your research team From their
com-ments, you may make changes and/or additions
to your overall concept map
One of the great benefits of concept mapping is that
it distinguishes between concepts that depend on one
another and ones which are distinct but related For
exam-ple, if you were to work out a concept map for
socioeco-nomic status (SES), you would certainly need to work in
qualitative and quantitative factors that indicate social
sta-tus and those that indicate economic stasta-tus Income is part
of SES, so you would need some measure for that But you
wouldn’t say that income relates to SES, because they are
part of the same concept Many of my students,
recogniz-ing that racial categories relate to SES in the United States,
also try to fit race into their conceptualization But race is a
separate variable, one which can only be compared to SES
because the two are different things
The final concept map, as suggested previously, may
go through a series of further refinements as others review
the draft or as you review additional pieces of literature In
addition to the overall design of the research, you will also
need to consider other elements, including, for example,
the nature of the research setting and the appropriateness
of your subjects
2.6.3: Using a Concept Map
The concept map portrayed in Figure 2.2 provides a
graphic representation of a theorized set of relations in the
study of obesity and perceptions The map groups together
many ideas that pertain to the overall study into related
conceptual blocks The conceptual map suggests a series of
relationships and a series of processes that relate the parts
With this map, we can make concrete statements and offer
testable propositions about the social world in which
teen obesity is experienced It gives us tangible elements
to look at when making sense of complex phenomena A
concept map is therefore highly useful for theorizing and
predicting
The concept map also has a more technical use to
help in the research design By describing the
concep-tual processes that interest us, the map also tells us
what data we need to measure This data may be
bro-ken down into two parts: variables and relationships
For simplicity, let us consider a condensed concept
map relating two characteristics or events: A ➞ B This
representation tells us that we are theorizing that some
change in the state of whatever A is will cause a change
Trang 38Designing Qualitative Research 37
have opinions about the topics without actually possessing relevant knowledge or experience
Sometimes researchers identify what they believe
to be an appropriate study population, but they cannot immediately see where an appropriate setting might be for data collection For example, several years ago, Berg had a student interested in conducting a study about fear
of crime among the blind On the surface, this sounds like
a good research topic The problem arose when Berg asked him how (and where) he planned to access such a popula-tion of potential subjects
The student reported that he had discussed his need
to access blind people to conduct a study of their tions of fear of crime with another faculty member The faculty member—who was obviously not terribly versed
percep-in research methods—suggested that the student simply go
to one of the large introductory classes and divide the class
in half Then he suggested that the student have half the class place blindfolds over their eyes and spend a period
of time walking around campus (ushered by one of the other nonblindfolded students) Following this experience, the students could switch off, so both groups experienced
blindness Next, the class could be administered a
pencil-and-paper survey about their fear of crime, having now experienced the precariousness of not being able to see The student immediately recognized that this would not be an appropriate setting or sample for his study Wisely, how-ever, he did not argue with the faculty member but rather thanked him and explained that he wanted to conduct a more qualitative study (See “Disengaging,” in Chapter 6.)The student then did some background research and came up with an actual plan He indicated that he intended
to attend a summer camp for the blind sponsored by eral nonprofit agencies He had learned that the popula-tion of the camp came from the entire state and that no one who wanted to attend was ever turned away (those who could not afford to pay were awarded camp scholarships) Thus, the camp contained a population from various socio-economic strata, races, ages, and both men and women The student spent the summer and was able to conduct both nearly 60 interviews and some limited participant observation (Rounds, 1993) The quality of the research depended on the appropriateness of the research setting.Another problem may arise when one must pay a fee for accessing certain types of settings, such as oral history archives that may charge a fee for use, or a fee for reproduc-tion of various interviews (Ritchie, 2003) Most archives do not charge merely for a researcher examining materials they house, but may charge if the researcher plans on publishing long excerpts from material housed in the archive Cost may also become a factor for seemingly “public” settings, such as sporting events There was a time when a student researcher could go to all of their local professional team’s home games for a month to observe the crowd, but such
sev-available It is also wise to consider your own level of skill
and familiarity with the setting The study site or setting
should be a location where:
1 Entry or access is possible
2 The appropriate people (target population) are likely
to be available
3 There is a high probability that the study’s focuses,
and processes, people, programs, interactions, and
structures that are part of the research question(s) will
be available to the investigator
4 The research can be conducted effectively by an
indi-vidual or indiindi-viduals during the data-collection phase
of the study
Also, don’t use a shopping mall, unless your study is about
shopping Or, more generally, do not plan a research
proj-ect in some setting simply because the setting is available
Your research question is generally regarded as the
primary guide to the appropriate site or setting selection
(Flick, 2006; Leedy & Ormrod, 2004; Marshall & Rossman,
2006; Silverman, 2006) For example, if the research question
has to do with why some people remain in relationships
after having suffered domestic abuse, the data-collection
site must be a safe place related to these populations, such
as a shelter If you want to know more about women who
leave their husbands, try family court To understand
pop-ular representations of domestic abuse, look to television,
newspapers, and magazines Or, if your research concerns
effective intervention strategies, contact mediation centers
and family counseling services
In many cases, the decision to use a particular research
site is tied closely to obtaining access to an appropriate
population of potential subjects Poor study site selection
and poor sample decisions may weaken or ruin eventual
findings (National Research Council, 2005) You must be
careful to identify an appropriate population, not merely an
easily accessible one For instance, let’s say you wanted to
conduct a study investigating the opinions or practices of
Native Americans One easy way of locating a site and
pop-ulation might be to turn to college students After all,
col-lege students are easy to locate on colcol-lege campuses They
are likely to be willing to take part in an interview—either
out of curiosity or to help out another student But you must
ask the question: What pertinent information will the
aver-age non–Native American college student have regarding
how Native Americans think, perceive their social world,
or practice their particular lifestyles? In other words, if you
want to know about Native Americans, then you need to
locate a setting where Native Americans can be accessed
This point may seem obvious, but we still encounter studies
of the impact of new policies that are conducted among
vot-ers who don’t actually have any knowledge of the policy, or
studies of race relations that only gather data from members
of one racial group The subjects in these groups may well
www.downloadslide.com
Trang 39populations can be described, but not enumerated or listed Some populations are deliberately hidden, while others may simply be difficult to locate In these situa-
tions, investigators rely on nonprobability samples.
In nonprobability sampling, the investigator does not base his or her sample selection on probability theory Rather, efforts are undertaken (1) to create a kind of quasi-random sample and (2) to have a clear idea about what larger group
or groups the sample may reflect Nonprobability samples offer the benefits of not requiring a list of all possible ele-ments in a full population and the ability to access otherwise highly sensitive or difficult-to-research study populations For example, it would be very difficult to undertake a study
of active prostitutes, drug users, or shoplifters because those are illegal activities and relatively hidden It would be virtu-ally impossible to create a list of all of the petty criminals in
a given area At best, one might acquire a listing of all the known (convicted) members in these groups if the study has the participation of a law enforcement agency, though that participation may discourage the cooperation of those who have engaged in illegal activities Thus, frequently in the social sciences, a researcher is presented with interesting and potentially important research questions that cannot be answered by a probability sampling technique From the perspective of qualitative research, nonprobability sampling tends to be the norm The following sections describe the four most common types of nonprobability samples
ConvEniEnCE SamplES The convenience sample is
sometimes referred to as an accidental or availability sample
(Babbie, 2007; Mutchnick & Berg, 1996; Polit & Beck, 2007) This category of sample relies on available subjects—those who are close at hand or easily accessible For example, it
a study now would require a large grant Most bars don’t
charge for entry, but costs can add up for a researcher
try-ing to “participate” in the setttry-ing Situations such as these
require one to reconcile the benefits of the particular source
materials (the particular archive or location as a research
setting), possible alternative settings, and the costs
2.6.5: Sampling Strategies
The logic of using a sample of subjects is to make
infer-ences about some larger population from a smaller one—the
sample Such inferences succeed or fail according to how
well the sample represents the population In large
quantita-tive studies, the investigator is keenly concerned with
prob-ability sampling The concept of probprob-ability sampling is based
on the notion that a sample can be selected that will
math-ematically represent subgroups of some larger population
(Shaughnessy, 2008; Vito, Kunselman, & Tewksbury, 2008)
The parameters required for creating these probability
sam-ples are quite restrictive but allow the investigator to make
various inferential hypothesis tests (using various statistical
techniques) The most commonly discussed probability
sam-ple is the simsam-ple random samsam-ple The simsam-ple random samsam-ple
most closely approximates the ideals in probability sampling
To accomplish a simple random sample, each element in the
full population must have an equal and independent chance
of inclusion in the eventual sample to be studied Simple
ran-dom sampling typically begins with a full listing of every
ele-ment in the full population to be investigated (see Figure 2.3)
The social sciences often examine research
situa-tions in which one cannot select the kinds of probability
samples used in large-scale surveys and which conform
to the restricted needs of a probability sample Many
Figure 2.3 Probability Sampling Strategies
Simple Random Sampling Typically, this procedure is intended to produce a representative sample The
process draws subjects from an identified population in such a manner that every unit in that population has
precisely the same chance (probability) of being included in the sample.
Systematic Random Sampling The use of a systematic sample provides a convenient way to draw a
sample from a large identified population when a printed list of that population is available In systematic
sampling, every nth name is selected from the list Usually the interval between names on the list is
determined by dividing the number of persons desired in the sample into the full population For example, if a
final sample of 80 was desired and the population list contained 2,560 names, the researchers would divide
2,560 by 80 The resulting 32 becomes the interval between names on the list It is important, however, to
begin the list at some random starting place Frequently, researchers select a number between 1 and 20
(usually taken from a random numbers table) and begin at that location on the list and then stop at every nth
name—in our example, at every thirty-second name on the list.
Stratified Random Sampling A stratified sample is used whenever researchers need to ensure that a
certain sample of the identified population under examination is represented in the sample The population
is divided into subgroups (strata), and independent samples of each stratum are selected Within each
stratum, a particular sampling fraction is applied in order to ensure representativeness of proportions in the
full population Thus, sampling fractions in some strata may differ from those of others in the same sample
Stratified samples can be used only when information is available to divide the population into strata Data
Storage Retrieval and Analysis.
Trang 40Designing Qualitative Research 39
other people who possess the same relevant attributes they do—in effect, a chain of subjects driven by the referral of one respondent of another The process is continued until a sufficient number of participants have been recruited
Quota SamplES A quota sample begins with a kind of matrix or table that creates cells or strata The researcher may wish to use gender, age, education, or any other attributes
to create and label each stratum or cell in the table Which attributes are selected will depend on the research question
on which the study focuses The quota sampling strategy then uses a nonprobability method to fill these cells Please note that I said “method,” not “methods.” Each category in the overall sample must be filled using the same recruitment strategy in order for the resulting groups to be comparable Next, the researcher needs to determine the proportion of each attribute in the full-study population (Babbie, 2007) For instance, let’s say a researcher wants to study perceptions of violence among people in the United States, with a special interest in people over age 65 Census data would provide the researcher with reasonable estimates of the percentage
of the population over age 65, as well as various categories under age 65 The research could create various age co-horts—people over 65, 45–65, 25–44, and under 25 Next, the researcher could determine the proportion of people in each
of these age groups Following this, the investigator could select a region of the country and sample people in that area, identifying the same proportion of people for each age co-hort as identified in the census data For quota sampling, one may place ads or otherwise cast a large net to bring people
of all ages into the study However, as each category fills its quota, subsequent volunteers in those categories will be turned away The recruitment period overall would remain open until all of the categories have reached their quotas.2.6.6: Representativeness
The examples given here used the most basic and visible characteristics to differentiate among members of a popula-tion It’s easy to see the value in recruiting equal numbers
of men and women for a study in which we think gender would be a factor in the results Other studies naturally stratify among different age groups or socioeconomic sta-tuses But demographics are not always the useful catego-ries For example, a study of interorganizational relations among groups in a protest movement would need to include both formally hierarchical organizations and those practic-ing participatory democracy A study of the use of dance therapy among trauma victims does not obviously need an equal number of black-and-white participants, but it might need subpopulations distinguished by categories of trauma Study subjects, whether individual or corporate, need to rep-resent some experience, phenomena, or characteristics that are pertinent to the research question It is the question, not the demographics, that determines representativeness
used to be fairly common for college and university
pro-fessors to use their students as subjects in their research
projects This technique has been used all too frequently and
has some serious risks associated with it (Gilligan, 1982)
Specifically, often a researcher is interested in studying
characteristics or processes that college students simply are
not equipped to offer information about Consider again, for
example, the suggested use of blindfolded students
Under certain circumstances, this strategy is an
excel-lent means of obtaining preliminary information about some
research question quickly and inexpensively For example,
if an investigator were interested in examining how college
students perceive drinking and drunkenness, he or she could
easily make use of a convenience sample of college students
If, on the other hand, the researcher was interested in
study-ing self-images among blue-collar workers, he or she could
not use this convenience sample of college students and
simply ask them to pretend that they are blue-collar workers
when answering the researcher’s questions In other words,
convenience samples must be evaluated very carefully for
their appropriateness of fit for a given study
purpoSivE SamplES This category of sampling is
sometimes called judgmental sampling (Hagan, 2006) When
developing a purposive sample, researchers use their
spe-cial knowledge or expertise about some group to select
subjects who represent this population In some instances,
purposive samples are selected after field investigations on
some group in order to ensure that certain types of
individ-uals or persons displaying certain attributes are included in
the study Despite some serious limitations (e.g., the lack of
wide generalizability), purposive samples are often
profit-ably used by researchers Laquinta and Larrabee (2004), for
example, used a purposive sample to examine the lived
experiences of a small sample of patients with rheumatoid
arthritis The results were a rich and textured description of
what it is like to live with rheumatoid arthritis, as well as a
number of nursing practices related to care quality
Snowball SamplES Another nonprobability
sam-pling strategy, which some may see as similar to
conve-nience sampling, is known as snowball sampling, chain
referral sampling (Biemacki & Waldorf, 1981; Owens, 2005;
Penrod, Preston, Cain, & Stark, 2003), or
respondent-driven sampling (Heckathorn & Jeffri, 2003) Snowballing
is sometimes the best way to locate subjects with certain
at-tributes or characteristics necessary in the study Snowball
samples are particularly popular among researchers
in-terested in studying various classes of deviance, sensitive
topics, or difficult-to-reach populations (Lee, 1993)
The basic strategy of snowballing involves first
iden-tifying several people with relevant characteristics and
interviewing them or otherwise gathering data from them
These subjects are then asked for the names (referrals) of
www.downloadslide.com